United States

Department
of Agriculture
Research and Cumulative
Forest Service
Watershed Effects
Pacific Southwest
Research Station

General Technical
Report PSW-GTR-141 Leslie M. Reid
Reid, Leslie M. 1993. Research and cumulative watershed effects. Gen. Tech. Rep. PSW-GTR-
141. Albany, CA: Pacific Southwest Research Station, Forest Service, U.S. Department of
Agriculture; 118 p.

The mandate for land managers to address cumulative watershed effects (CWEs) requires that
planners evaluate the potential impacts of their activities on multiple beneficial uses within the
context of other coexisting activities in a watershed. Types of CWEs vary with the types of land-
use activities and their modes of interaction, but published studies illustrate both descriptive and
predictive evaluations of many of these types. Successful evaluations have generally used geomor-
phological and ecological approaches based on the understanding of the processes involved. In
contrast, most generalized “cookbook” analysis procedures are shown to be unable to assess
accumulations of impacts through time, usually cannot evaluate the range of activities and uses that
are necessary, and are rarely validated. A general approach to evaluation is proposed, and the types
of information available for assessments are reviewed.

Retrieval terms: watershed, cumulative impact, land-use planning, water quality

The Author:

Leslie M. Reid is a research geologist with the Station’s Hillslope Processes/Fisheries Research
Unit, and is stationed at the Redwood Sciences Laboratory, 1700 Bayview Drive, Arcata, CA 95521.

Foreword and Acknowledgments:

This report was prepared under cooperative agreement PSW-88-CL-009 (CDF 8CA74571)
with the California Department of Forestry and Fire Protection, P.O. Box 944246, Sacramento, CA
94244-2460. This version differs from the final report to the Department primarily in editing
details, omission of two appendices, addition of two figures, revisions to tables, and inclusion of
more detail about cited studies. Trade names and commercial enterprises are mentioned solely for
information and do not imply the endorsement of those products or organizations. This publication
neither recommends any pesticide uses reported nor implies that they have been registered by
appropriate government agencies.
John Munn, Peter Cafferata, Jennifer Nielsen, and Larry Costick provided extremely helpful
reviews of the manuscript, and innumerable scientists and land managers (listed, in part, in the
Appendix) gave freely of their time for lengthy discussions of ideas, research, and the challenges
confronting conscientious land management.

Note: This publication was reprinted December, 1998 from a reconstructed layout. It closely
resembles the original printing, but a difference in text style has caused a slight shift in pagination.
The Table of Contents has been revised to reflect the changes. Also, Figure 4, (previously on page
29) now appears on page 28, and Tables 5 (previously on page 28) appears on page 29.

Publisher:

Pacific Southwest Research Station

Albany, California
(Mailing address: P.O. Box 245, Berkeley, CA 94701-0245
Telephone: 510-559-6300)

March 1993
Research and Cumulative Watershed Effects
Leslie M. Reid

Contents

List of Figures ....................................................................................................................................... v

List of Tables ........................................................................................................................................ v
Glossary of Acronyms ......................................................................................................................... vi
In Brief ................................................................................................................................................ vii
1. The Problem of Cumulative Watershed Effects ......................................................................... 1
Definitions ................................................................................................................................... 1
History of the Issue ..................................................................................................................... 3
The Major Issues ......................................................................................................................... 4
Goals of the Review .................................................................................................................... 4
Scope of the Review .................................................................................................................... 4
2. How Cumulative Watershed Effects Occur ................................................................................ 6
Watershed Function .................................................................................................................... 6
Runoff Generation ............................................................................................................. 6
Channels ............................................................................................................................. 6
Sediment Production .......................................................................................................... 6
Interactions Between Hillslopes and Channels ................................................................ 7
Physical Basis for Cumulative Watershed Effects ........................................................... 7
Ecosystem Function .................................................................................................................... 7
Constraints on Individuals and Species ............................................................................ 8
Habitat ................................................................................................................................ 8
Biological Communities .................................................................................................... 8
Response to Change ........................................................................................................... 9
Biological Basis for Cumulative Watershed Effects ........................................................ 9
A Framework for Understanding CWEs .................................................................................. 10
Issues that Complicate the Understanding of CWEs ............................................................... 11
3. Research Methods for Understanding Cumulative Watershed Effects ................................. 12
Requirements of CWE Research .............................................................................................. 12
Research Approaches and Methods .......................................................................................... 12
Descriptive Studies .......................................................................................................... 12
Extensive Surveys ............................................................................................................ 13
Experimentation ............................................................................................................... 14
Modeling .......................................................................................................................... 15
Monitoring ....................................................................................................................... 16
Statistics ........................................................................................................................... 16
Scale and Focus of Studies ....................................................................................................... 17
Process Studies ................................................................................................................ 17
Watershed Case Studies .................................................................................................. 17
Paired- and Multi-Basin Studies ..................................................................................... 17
The System as Focus ....................................................................................................... 18
 4. Assessment of Cumulative Watershed Effects .......................................................................... 19
Types of Cumulative Watershed Effects ............................................................................... 19
On-Site and Off-Site CWEs .......................................................................................... 19
Interactions that Generate CWEs .................................................................................. 19
Retrospective Assessments of Cumulative Watershed Effects ............................................. 20
Examples of Retrospective Analyses ........................................................................... 20
Same-Influence Effects .......................................................................................... 20
Complementary Influences .................................................................................... 22
Cascading Influences ............................................................................................. 23
Interdependent Influences ...................................................................................... 23
Patterns of Retrospective Understanding ..................................................................... 24
Prediction of Cumulative Watershed Effects ......................................................................... 25
Examples of Predictive Studies .................................................................................... 25
Same-Influence Effects .......................................................................................... 25
Complementary Influences .................................................................................... 26
Cascading and Interdependent Influences ............................................................. 27
Patterns of Predictive Understanding ........................................................................... 27
General Methods for Predicting Cumulative Watershed Effects .......................................... 27
Examples of General Predictive Methods .................................................................... 28
Equivalent Clearcut Area ....................................................................................... 28
Klock Watershed Cumulative Effects Analysis .................................................... 30
Equivalent Roaded Area ........................................................................................ 30
R-1/R-4 Sediment-Fish Model .............................................................................. 31
California Department of Forestry Questionnaire ................................................ 31
Water Resources Evaluation of Non-point Silvicultural Sources ........................ 32
Limiting Factor Analysis ....................................................................................... 32
Rational Approach (Grant) .................................................................................... 33
Discussion of General CWE Assessment Procedures .................................................. 33
Methods Under Development ....................................................................................................... 35
Sequoia National Forest Method ............................................................................................ 35
TFW Method ........................................................................................................................... 35
NCASI Method ....................................................................................................................... 35
Prospects for a General CWE Evaluation Procedure ............................................................ 35
Attributes Required of a General Method .................................................................... 36
Development and Use of a General Method ................................................................ 36
5. Effects of Land-Use Activities on Environmental Parameters ............................................... 38
Roads ....................................................................................................................................... 38
Roads and Vegetation ................................................................................................... 38
Roads and Soils ............................................................................................................. 39
Roads and Topography ................................................................................................. 39
Roads and Chemicals .................................................................................................... 39
Impoundments and Water Development ................................................................................ 39
Timber-Management Activities ............................................................................................. 40
Timber Management and Vegetation ........................................................................... 40
Timber Management and Soils ..................................................................................... 41
Timber Management and Topography ......................................................................... 42
Timber Management and Chemicals ............................................................................ 42
Indirect Effects of Timber Management on Environmental Parameters ..................... 42
Grazing .................................................................................................................................... 43
Grazing and Vegetation ................................................................................................ 43
Grazing and Soils .......................................................................................................... 45
Grazing and Topography .............................................................................................. 45
Other Effects of Grazing ............................................................................................... 45
Indirect Effects of Grazing on Environmental Parameters .......................................... 45
Interactions Between Grazing and Timber Management ............................................ 46

ii USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Mining ..................................................................................................................................... 46
Mining and Vegetation .................................................................................................. 46
Mining and Soils ........................................................................................................... 46
Mining and Topography ................................................................................................ 47
Mining and Chemicals .................................................................................................. 47
Interactions Between Mining and Timber Management ............................................. 47
Agriculture .............................................................................................................................. 47
Urbanization ............................................................................................................................ 48
Urbanization and Vegetation ........................................................................................ 49
Urbanization and Chemicals ......................................................................................... 49
Other Effects of Urbanization ....................................................................................... 49
Interactions Between Urbanization and Timber Management .................................... 50
Flood Control and Navigation ................................................................................................ 50
Recreation and Fishing ........................................................................................................... 51
General Patterns of Change .................................................................................................... 53
Cumulative Effects of Land Use on Environmental Parameters ........................................... 54
Research Needs for Understanding Changes in Environmental Parameters ........................ 54
Understanding Environmental Parameters ................................................................... 55
Understanding the Effects of Land Use ........................................................................ 54
6. Effects of Land Use and Environmental Change on Watershed Processes .......................... 56
Relation Between Watershed Processes and CWEs .............................................................. 56
Effects of Environmental Change on Hydrology .................................................................. 57
Water Input .................................................................................................................... 57
Runoff Generation ......................................................................................................... 60
Water Transport on Hillslopes ...................................................................................... 62
Water Transport in Channels ........................................................................................ 62
Water Budgets and Runoff Modeling ........................................................................... 63
Effects of Environmental Change on Sediment ..................................................................... 64
Erosion and Sediment Transport on Hillslopes ............................................................ 64
Surface Erosion ...................................................................................................... 64
Gullies .................................................................................................................... 65
Landslides ............................................................................................................... 65
Soil Creep ............................................................................................................... 66
Treethrow ............................................................................................................... 66
Sediment Delivery to Streams ...................................................................................... 66
Erosion, Transport, and Deposition in Channels ......................................................... 66
Bank and Channel Erosion .................................................................................... 66
Channel Sediment Transport ................................................................................. 67
Sediment Deposition in Channels ......................................................................... 68
Land Use and Sediment Yield ...................................................................................... 69
Sediment Budgeting and Modeling .............................................................................. 69
Effects of Environmental Change on Organic Material ........................................................ 72
Production of Organic Material from Streamsides ...................................................... 73
In-Channel Production of Organic Material ................................................................. 74
Modification of Organic Material ................................................................................. 74
In-Channel Transport of Organic Material ................................................................... 74
Organic Material Budgets ............................................................................................. 75
Effects of Environmental Change on Chemicals ................................................................... 75
Input and Removal of Chemicals ................................................................................. 75
Transport of Chemicals ................................................................................................. 76
Alteration of Chemicals ................................................................................................ 77
Accumulation of Chemicals .......................................................................................... 77
Nutrient Cycling, Chemical Budgets, and Modeling ................................................... 77
Effects of Environmental Change on Heat ............................................................................ 78
Research Needed to Understand Watershed Processes ......................................................... 79

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. iii

 7. Impacts on Beneficial Uses and Values ...................................................................................... 81
Impacts on Fisheries Resources and Aquatic Communities .................................................... 81
Flow Characteristics ........................................................................................................ 81
Channel Morphology ....................................................................................................... 82
Water Temperature .......................................................................................................... 83
Water Quality ................................................................................................................... 84
Food Resources ................................................................................................................ 85
Predation and Fishing Pressure ....................................................................................... 85
Other Effects .................................................................................................................... 86
Impacts on Timber Resources ................................................................................................... 86
Loss of Land Base ........................................................................................................... 86
Forest Productivity and Mortality ................................................................................... 86
Impacts on Recreation and Esthetic Values ............................................................................. 87
Recreational Needs .......................................................................................................... 87
Physical Site Characteristics ........................................................................................... 88
Crowding and Ambiance ................................................................................................. 88
Impacts on Water Supply and Power Generation .................................................................... 89
Water Quality ................................................................................................................... 90
Water Quantity and Runoff Timing ................................................................................ 90
Water Storage .................................................................................................................. 90
Impacts on Floodplain and Channel Use .................................................................................. 90
Stream Flow ..................................................................................................................... 90
Channel Morphology ....................................................................................................... 91
Other Changes .................................................................................................................. 91
Impacts on Other Resources and Values .................................................................................. 91
Research Needed to Understand Impacts ................................................................................. 92
8. Summary of Research and Cumulative Watershed Effects .................................................... 93
A Framework for Understanding CWEs .................................................................................. 93
State of Present Knowledge ...................................................................................................... 93
Methods for Evaluating CWEs ................................................................................................. 94
A Framework for CWE Evaluation .......................................................................................... 95
The Role of Research in Assessing CWEs ............................................................................... 95
Appendix: People Interviewed During Report Preparation ....................................................... 96
References .......................................................................................................................................... 97

iv USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

List of Figures

1. Types of System Response ....................................................................................................................................................... 2

2. Framework for Understanding Cumulative Watershed Effects ............................................................................................ 10
3. Combinations of Influences That Can Generate CWEs ........................................................................................................ 20
4. Cumulative Effect of Differing Recovery Times for a Driving Variable and an Impact .................................................... 28
5. Influences That Generate On-Site and Off-Site CWEs ......................................................................................................... 93

List of Tables

1. Advantages and Disadvantages of Research Approaches and Methods .............................................................................. 13

2. Examples of Studies Diagnosing Cumulative Effects .......................................................................................................... 21
3. Approaches Used by Studies Described in Table 2 .............................................................................................................. 24
4. Examples of Predictive Cumulative Effect Studies .............................................................................................................. 25
5. General Procedures for Evaluating CWEs ............................................................................................................................ 29
6. Attributes of CWE Evaluation Procedures ............................................................................................................................ 34
7. Potential Direct Effects of Various Land-Use Activities on Watershed Properties ............................................................ 52
8. Interactions Between Altered Environmental Parameters .................................................................................................... 53
9. Effects of Altered Environmental Parameters on Watershed Processes .............................................................................. 56
10. Interactions Between Watershed Processes ........................................................................................................................... 57
11. Studies of the Hydrologic Effects of Roads and Timber Management ................................................................................ 58
12. Studies of Sediment Production from Roads and Timber Management .............................................................................. 70
13. Sediment Budgeting Studies .................................................................................................................................................. 72

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. v

Glossary of Acronyms

BKD Bacterial kidney disease

BMP Best management practice
CDF California Department of Forestry and Fire Protection
CEQ Council on Environmental Quality
CREAMS Chemicals, Runoff, and Erosion in Agricultural Management Systems
CWE Cumulative watershed effect
ECA Equivalent Clearcut Area
ERA Equivalent Roaded Area
GIS Geographic information system
LFA Limiting Factor Analysis
KWCEA Klock Watershed Cumulative Effects Analysis
NCASI National Council of the Paper Industry for Air and Stream Improvement
ORV Off-road vehicle
PCB Polychlorinated biphenyl
PRMS Precipitation Runoff Modeling System
RA Rational Approach (Grant)
R-1 Region 1 of the USDA Forest Service
R-4 Region 4 of the USDA Forest Service
RV Recreational vehicle
TFW Timber/Fisheries/Wildlife group
THP Timber harvest plan
TOC Threshold of concern
USGS U.S. Geological Survey
USLE Universal Soil Loss Equation
WEPP Water Erosion Prediction Project
WRENSS Water Resources Evaluation of Non-point Silvicultural Sources

vi USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

In Brief . . .
Reid, Leslie M. 1993. Research and cumulative watershed effects.
Gen. Tech. Rep. PSW-GTR-141-WWW. Albany, CA: Pacific
Southwest Research Station, Forest Service, U.S. Department
of Agriculture; 118 p.
Retrieval terms: watershed, cumulative impact, land-use planning
C umulative watershed effects (CWEs) include any changes
that involve watershed processes and are influenced by
multiple land-use activities. CWEs do not represent a new
type of impact, and almost all land-use impacts can be
evaluated as CWEs. The CWE concept is important primarily
because it identifies an approach to impact evaluation and
mitigation that recognizes multiple influences. The
significance of a CWE varies with the type of resource or
value impacted and is determined on political, economic, and
cultural grounds. In contrast, impact magnitude can be assessed
objectively by measuring physical and biological changes.
Most CWEs are incremental results of multiple controlling
factors; rarely can a single threshold value be identified for
provoking a response. (Chapter 1)
Watersheds are topographical forms that concentrate runoff.
They are sculpted by production and transport of water and
sediment. These media transport other watershed products,
such as organic material, chemicals, and heat. If one watershed
process or product is altered by land use, others change in
compensation. Changes often influence or are influenced by
biological communities, and most biological changes have
repercussions throughout a biological community. Both
physical and biological systems continually undergo change
even in an undisturbed state. (Chapter 2)
CWEs are caused by changes that accumulate in time or
space. Land use directly affects only a small number of
environmental parameters, including vegetation, soil properties,
topography, chemicals, and fauna. These parameters, in turn,
influence production and transport of water, sediment, organic
matter, chemicals, and heat. Onsite impacts can occur if the
triggering change or the resulting impact is persistent in time.
Off-site impacts can occur only if watershed processes or
products are altered, because something must be transported
for a remote effect to occur. (Chapter 2)
The manifestation of CWEs is complicated by lags in
system response to change, geographic decoupling between
cause and effect, site-specific variations in impact expression,
accumulation of innocuous changes to the point that a
catastrophic change is triggered, the ability of high-magnitude
events such as storms or earthquakes to trigger delayed
impacts, and interaction between changes that modify their
expression. (Chapter 2)
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
CWE evaluation must be based on the basic understanding
of watershed and ecological processes already provided by
studies in hydrology, geomorphology, forestry, ecology,
and many other fields. These studies have employed a variety
of approaches, including qualitative description, statistical
comparison, monitoring, experimentation, and modeling.
Each approach has advantages and disadvantages, and
methods must often be combined to solve a particular
problem. Studies also vary in their selected focus: some
concentrate on the mechanics of an isolated process, while
others compare multiple sites or watersheds through time.
Research that specifically addresses CWEs usually must
consider large temporal and spatial scales and often needs to
include interdisciplinary work. (Chapter 3)
Cumulative effects can be caused by repeated, progressive,
sequential, and coexisting land-use activities. They can occur
because of a single type of influence on an environmental
parameter (for example, many types of land use can compact
soils), complementary influences (for example, both increased
compaction and altered snow accumulation can affect flood
peaks), cascading influences (one type of land use can
influence a second to cause an impact, as when urbanization
increases recreational pressure, increasing trail erosion),
and interdependent influences (for example, two introduced
chemicals can react to produce a third). Many studies
document the occurrence of such impacts, and a few attempt
to predict them at particular sites. (Chapter 4)
Several methods have been developed to evaluate potential
CWEs from particular land-use activities. However, methods
rarely address accumulation of effects through time, few are
adequately validated, and rarely do they address more than
one type of land use or impact. Approaches fall into three
categories: procedures for calculating values or indices
(including the Equivalent Roaded Area, Equivalent Clearcut
Area, and Region-1/Region-4 methods), a collection of
analytic procedures (the WRENSS procedures), and a
checklist of issues to consider during evaluation (the method
developed by the California Department of Forestry and
Fire Protection). None is adequate for a complete CWE
analysis. (Chapter 4)
A valid general approach to CWE analysis would be
capable of assessing the full variety of impacts and land-use
activities at all potentially impacted sites. To remain credible,
it would need to incorporate new analytical methods as they
become available, and it would require comprehensive
validation and monitoring of component methods and results.
A useful format might use a CDF-style checklist to select
relevant analytical methods from a WRENSS-type collection
of procedures. Enough is already known of most environmental
vii
parameters, watershed processes, and impacts to develop
preliminary procedures. (Chapter 4)
The effects of land use on specific environmental parameters
have been widely documented, and a selection of these is
described in Chapter 5. Modification of environmental
parameters, in turn, provokes changes in watershed processes.
viii
These influences have also been widely studied (Chapter 6).
The impact of altered environmental parameters and watershed
processes on particular uses and values is more poorly
understood (Chapter 7). Research like that described in
Chapters 5, 6, and 7 provides the foundation for future methods
of evaluating CWEs.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Chapter 1
The Problem of Cumulative Watershed Effects
C umulative watershed effects (CWEs) present a potent
challenge to land managers. Land must now be man-aged
so that the combination of activities within a watershed does
not significantly impact other beneficial uses. Managers are
required to evaluate the interaction of their activities with
those of the past, present, or future, and to assess their combined
effect on other existing or potential uses or values. Even
though the effect of an activity might be minimal when
considered in isolation, it may combine with the effects of
adjacent or future projects to cause unacceptable impacts.
Meanwhile, the definition of “acceptable” is becoming narrower.
Land managers must be able to predict the environmental
effects of planned activities if they are to avoid CWEs.
Before prediction is possible, however, we need to understand
how CWEs come about, how they are expressed, and how
systems respond to them. The answers to these questions lie
in an understanding of how physical and biological systems
work, and this is also the general goal of science.
This report outlines issues underlying CWE concerns,
discusses mechanisms by which CWEs are expressed,
describes methods used in their analysis, and summarizes
research relevant to their evaluation.
Definitions
The definition of CWEs provided a controversial issue for
conferences and workshops of the late 1970s and early 1980s.
In 1971, the Council on Environmental Quality (CEQ) agreed
on a restatement of the principles expressed in the National
Environmental Policy Act of 1969 (CEQ Guidelines, 40
CFR 1508.7, issued 23 April 1971):
“Cumulative impact” is the impact on the
environment which results from the incremental
impact of the action when added to other past,
present, and reasonably foreseeable future actions
regardless of what agency (Federal or non-
Federal) or person undertakes such other actions.
Cumulative impacts can result from individually
minor but collectively significant actions taking
place over a period of time.
This definition still left room for interpretation. Were additive
impacts CWEs? Or did impacts have to be synergistic, such that
the effect of two activities is more than the sum of their independent
effects? Were decrementally synergistic changes included?
Acceptance of the additive interpretation would allow CWEs
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
to include the effects of any combination of uses, and this was
considered by some to undermine the intent of the definition.
They feared that cumulative effects would become simply a
synonym for land-use effects, because no land-use activity is
carried out in isolation. “Too broad,” the synergists complained.
However, breadth appears to be an intent of the definition,
and was certainly a goal of those driving the legal process.
The CEQ definition is useful not in circumscribing a certain
class of environmental impacts, but in identifying an approach
to land management and impact mitigation that had not been
employed in the past. Over time, the literal version of the CEQ
definition has been accepted by most researchers and jurists.
Impact assessment has traditionally been carried out on a
project scale. In the case of timber resources, this approach
has taken the form of best management practices (BMPs).
The BMP approach is based on the premise that if on-site
effects of a project are held to an acceptable level, then the
project is acceptable, regardless of activities going on around
it. Interactions between projects are beyond the scope of
BMP analysis, and operational controls are applied only to
individual projects. In addition, BMPs are designed to be
“practical”: the interests of the impactee and impactor are
balanced in such a way as to make the remedy economically
palatable to the impacting user, rather than having it be
based primarily on the needs of the impacted resource.
In this context, the CEQ definition was revolutionary and
lent legal credence to injuries sustained by other uses. A
land-use activity would now be judged according to its
contribution to the overall impact from all activities in a
watershed. Efforts to narrow the definition of CWEs thus
skirt an important aspect of the cumulative effects concept
by concentrating on the form of impact rather than on the
philosophical approach to impact evaluation. In addition,
land managers do not have the luxury of limiting the problem’s
scope by constraining the definition. Injured parties will
continue to sue when damages can be attributed to activities
taking place in their watersheds, and courts will continue to
apply the legal definition.
For the purposes of this paper, a cumulative effect is any
environmental change influenced by a combination of land-
use activities. Redd siltation affected by accelerated
landsliding in a managed basin is thus considered a cumulative
effect, whether the preponderance of sediment is natural or
management-induced. In contrast, individual effects can be
traced to a specific activity at a specific site, and high
sediment levels in gravels immediately downstream of an
isolated landslide might be evaluated as an individual effect.
However, if the landslide resulted from a combination of
activities (e.g., oversteepening of a slope by excavation,
combined with an altered soil moisture regime caused by
1
logging), or altered runoff promoted sediment accumulation
from the slide, or sedimentation combined with other impacts
to contribute to a decline in salmon populations, then even
this well-constrained impact may be considered either a
CWE or a part of one. Almost all effects of land use can be
evaluated as cumulative effects because most are influenced
by more than one aspect of land use. In addition, air pollution
and other systemic changes are present at most sites and thus
interact with other use-related changes.
There is also disagreement about the meaning of
“watershed” in “cumulative watershed effect.” Some interpret
this to include any changes occurring within the bounds of a
drainage basin, so the watershed is simply the location of an
impact and does not necessarily play a role in its generation.
By this definition, decreased habitat for narrowly ranging
species might be considered a CWE, while similar decreases
for species whose individuals range beyond the watershed
are not. This interpretation produces inconsistencies resulting
from watershed scale. For example, fragmentation of spotted
owl habitat is neither meaningful nor assessable in a 5-ha
headwater catchment, but may be extremely important in a
large river basin.
Others interpret “watershed effects” to include only those
changes occurring to resources influenced directly or indirectly
by watershed processes, so processes of water and sediment
transport are functionally linked to the expression of impacts.
To provide a consistent definition that avoids problems of
watershed scale, this definition will be used in this paper.
Definition of the “significance” of particular CWEs is
also a focus of controversy, and Thompson (1990) reviewed
methods for evaluating impact significance. Some aspects of
significance can be defined economically. If oyster beds are
damaged, for example, costs and damages sustained by all
dependents of the oyster industry, including oyster farmers,
wholesalers, service industries, restaurant patrons, and others,
can be summed to estimate economic impact.
However, impacts on noneconomic values must also be
evaluated. Some resources have widely recognized intrinsic
and cultural worth that cannot be translated into dollars:
What price tag can be attached to potential medical discoveries
from the Pacific yew? How much is the esthetic integrity of a
trout stream worth? And how does one compensate in dollars
for cultural losses incurred by inundation of a Native American
ceremonial site? These impacts cannot be easily expressed as
economic effects on the primary users or on the public as a
whole, yet they are of concern to many who will never see or
experience them. If the Arctic National Wildlife Refuge were
valued only by those likely to visit it, use of that land would
not be controversial. Value in such cases is culturally defined,
so significance is a political or cultural issue rather than a
scientific or economic one. Economists have only recently
begun to develop methods of assessing “existence” values
(e.g., Pope and Jones 1990, Sanders and others 1990).
The significance of identical changes differs with the
locale and types of beneficial uses present, because
significance of an impact is defined from the point of view
2 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
of the resource or value affected. For example, a high sediment
load might be identified as a problem because it increases
water treatment costs in a community, but the same impact
might go unnoticed if the community were not present.
Similarly, a change can be beneficial to some uses and
harmful to others: gravel mines may benefit from the same
increases in sediment load that damage fisheries. Thus, a
CWE can be considered “good” or “bad” only with respect
to particular resources. Negative effects are most commonly
considered, both because positive effects do not motivate
lawsuits, and because a change from existing conditions
usually proves detrimental to at least some of the resources
that have equilibrated to the status quo.
Impact significance is sometimes evaluated against a “threshold
of concern,” where levels of impact below a certain value are
considered acceptable and higher levels are not. However, physical
and biological systems usually respond incrementally to change
(fig. 1a) rather than experiencing true thresholds (fig. 1b).
Even when a response involves a dramatic change in
state, factors causing the change rarely exhibit a simple
threshold. For example, an abrupt shift in channel pattern
rarely results from a single year’s sediment input; annual
variation in sediment input under undisturbed conditions
would likely have included values characteristic of the altered
input rate (fig. 1c). Chronically increased inputs can increase
the probability of channel disruption, but there is rarely a
definable value at which response is certain.
Many sudden changes, such as landsliding or gully incision,
result from the interaction of multiple factors. For example,
the rain intensity capable of triggering slides varies with
antecedent moisture, soil depth, vegetation cover, and other
conditions that vary with time. As soon as conditions change,
the threshold rainfall intensity also changes.
A. INCREMENTAL NON-LINEAR B. THRESHOLD RESPONSE
RESPONSE
RESPONSE
RESPONSE
threshold
DRIVING VARIABLE DRIVING VARIABLE
C. RESPONSE TRIGGERED BY CHRONIC CHANGES
RATHER THAN INSTANTANEOUS ONES
DRIVING VARIABLE
response might
response might
TIME
Figure 1—Types of system response. “Response” may be percent
of channels gullied, smolt population, water treatments costs, and
so on. “Driving variable” may be sediment input, peakflow dis-
charge, length of active roads, or some other parameter.
 For these reasons, thresholds of concern can rarely be
equated to physically or biologically meaningful numbers.
Instead, a threshold must be defined as either a tolerable
level of injury to impacted resources or an acceptable
probability of detrimental response. It is the goal of research
to develop an understanding of CWEs, the physical and
biological processes driving them, and the physical and
biological processes they affect. Scientists are thus concerned
with quantifying the impacts, but judgments of significance
and tolerance are left to political and legal arenas.
History of the Issue
Cumulative effects have been with us since humans began to
manipulate their environment. Progressive salinization of
irrigated land contributed to the breakup of Sumerian civili-
zation by 1700 B.C. (Jacobsen and Adams 1958), and the
combination of agriculture, grazing, and logging in ancient
Greece may have promoted sedimentation that forced relo-
cation of ports (Kraft and others 1975). Deforestation and
land use in Great Britain is thought to have contributed to
transformation of forest land and pasture into peat bogs over
the past 5,000 years (Moore 1975), and the cumulative ef-
fects of this vegetation conversion are only now being re-
versed through reforestation. The British later became known
for their efforts to control cumulative impacts in their colo-
nies. In East Africa, for example, restrictions were put on
herd size, agricultural practices, and cutting of fuelwood,
often over the protests of indigenous peoples (e.g., Berry
and Townshend 1972).
Cumulative watershed effects were recognized in the
United States by the 1860s, when coarse sediment produced
by hydraulic mining in the California foothills began to
choke downstream channels, overrun productive agricultural
lands, and contribute to flooding of valley towns. These
impacts provoked legislation that banned hydraulic mining
and led to what may be the first rigorous CWE assessment.
In his 1917 report, “Hydraulic-mining debris in the Sierra
Nevada,” G.K. Gilbert tracked the movement of mining
sediment and evaluated its likely impact on navigation in
San Francisco Bay.
Gilbert explicitly described examples of CWEs. To explain
increased flood frequencies in the Central Valley, he cited
construction of levees in agricultural areas in combination
with aggradation of mining debris: “If these changes had been
independent of those wrought by mining debris they would
have resulted in the automatic deepening and widening of the
channels.” But because they occurred in combination, channel
flow capacities decreased and downstream cities were flooded.
Gilbert then evaluated the influence of mining debris on
growth of the San Francisco bay-mouth bar. His calculations
showed that reclamation of tidal marshes has a more profound
effect on bar sedimentation than the influx of mining debris,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
and he noted that “every acre of reclaimed tide marsh implies
a fractional reduction of the tidal current in the Golden Gate.
For any individual acre the fraction is minute, but the acres of
tide marsh are many, and if all shall be reclaimed, the effect at
the Golden Gate will not be minute.”
Gilbert was well-equipped to evaluate such problems
because, as a geomorphologist, he was experienced in dealing
with complicated, interacting systems. Geomorphologists
frequently must decipher complex causal relations, disentangle
interactions between multiple driving variables, and predict
the effects of environmental change. Typical
geomorphological problems include: What determines the
extent of channel networks? What controls channel form?
How does a channel adjust to a change in watershed
conditions? These questions are also fundamental to
understanding and predicting CWEs.
Ecology provides a biosciences analog to geomorphology.
The focus of ecological research is communities of interacting
organisms responding to each other and to changes in their
physical environment. As with geomorphology, ecology was
initially a descriptive science. It was not until the middle of
this century that process-oriented and experimental approaches
to ecology became widely practiced, resulting in insights that
interested a lay audience. This was the era of One Day on
Beetle Rock (Carrighar 1943), The Immense Journey (Eiseley
1946), and A Sand County Almanac (Leopold 1949). Soon
after came Silent Spring (Carson 1962), and capital-E Ecology
was suddenly part of the public consciousness. Interdependency
and ecological complexity became familiar ideas; noneconomic
values of wildlands were accepted as legitimate; and, within
a decade, CWEs became a recognized issue.
Thus, even though the term was recently coined, cumulative
watershed effects are not a new problem. Cumulative impacts
were suffered at least 4,000 years ago, and questions central
to an understanding of CWEs are the basis of sciences such
as ecology and geomorphology. The only “new” elements
are the legislative mandate to address the problem, the
widespread popular concern about it, and the social and
economic changes that are required to fulfill the legislative
directives.
Because of their ubiquity and complexity, CWEs have
been used as a legal tool for delaying controversial
management activities and as a generic reason for downstream
damages. A “solution” to the CWE problem requires a valid
method for predicting the environmental impacts of land use.
Such a solution would benefit all those interested in resources,
whether that interest be primarily as a resource user, a land
manager, or an environmentalist. If the costs and delays of
litigation were obviated, resource production costs would
decrease and budgets that now contribute primarily to
supporting the legal profession could be devoted to other
pressing needs. A valid method for predicting land-use effects
would allow selection of land-use plans that minimize impacts
while providing for truly sustainable industries. Impacts to
off-site users would be reduced, and the environment would
be better protected.
3

The Major Issues
Discussions and investigations of CWEs usually involve
several topics:
1. Do CWEs exist? The general question is easy to answer:
studies have described CWEs resulting from many land-use
practices, including forestry (Lyons and Beschta 1983),
urbanization (Hammer 1972; Leopold 1973), and mining
(Touysinhthiphonexay and Gardner 1984). In each case,
progressive land-use changes caused increasing impacts on
water quality, runoff volume, channel morphology, or habitat
quality. The question is harder to answer on the scale of
particular projects, and methods for detecting CWEs will be
examined in this review.
2. Which sites are susceptible to CWEs? Scientists have
long been interested in factors controlling a system’s response
to change. This research is directly applicable to the CWE
problem, and relevant studies will be described in this report.
3. How significant are the impacts? The magnitude of
changes caused by land use is widely measured, and such
studies will be discussed. But the significance of the damage
to a resource is often defined on social or political grounds,
and these will not be considered.
4. Who is responsible? CWEs result from multiple or
progressive activities that are often carried out by different
users. Effects are also complicated by the occurrence of
large storms or other natural events that make distinction
between natural changes and land-use impacts difficult.
Contributions to watershed response from different land
uses can be assessed using methods such as water and sediment
budgeting, and these will be discussed.
5. How can impacts be predicted? Several methods of assessing
CWEs have been developed and implemented, and their scientific
basis, application, and limitations will be described.
6. How can land be managed to avert or redress impacts?
This has been a major research focus for land management
agencies, and the answer requires a basic understanding of
how watersheds and ecosystems work. Once out of the realm
of basic understanding, however, the topic becomes one of
engineering design. Basic principles will be discussed, but
specific implementations will not.
Goals of the Review
Many CWE reviews are already available. Most are com-
pendia of symposium papers that either report research re-
sults (Callaham and DeVries 1987; NCASI 1984, 1986) or
outline perspectives on CWEs (Standiford and Ramacher
1981). Other volumes describe likely CWEs and review
relevant literature (Coats and others 1979, Geppert and oth-
4 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
ers 1984, Beanlands and others 1986), and some outline
needed research (Peterson and others 1987b, Sonntag and
others 1987).
The present review differs from these in several respects.
First, it describes many relevant studies in geomorphology,
hydrology, and ecology that have not been discussed elsewhere
because they do not overtly focus on CWEs. Inclusion of
these studies shows the types of background information
available for development of evaluation procedures and
indicates productive paths for future research. Similarly,
gaps and weaknesses in our overall understanding of watershed
function are revealed when published work is fitted into a
framework to describe watershed function. Second, this review
describes a variety of methods that are now being used to
evaluate cumulative watershed effects. Methods are critiqued
in the context of the working definition of cumulative
watershed effect, and a procedure for CWE evaluation is
proposed. Finally, much has changed since the most recent
reviews were published. New frameworks and methods for
understanding CWEs have been developed, and new
procedures for their evaluation are being implemented.
Scope of the Review
This report reviews research that adds to our understanding
of CWEs, whether or not it was specifically designed to
address them. Much of our understanding of CWE mecha-
nisms emerged from basic research in the fields of hydrol-
ogy and geomorphology. Resources affected by change are
often biological, so ecological investigations also produce
fundamental information on how and why impacts are ex-
pressed. Chapter 2 provides an overview of watershed and
ecosystem function and describes CWE mechanisms. CWE
evaluation can be simplified by understanding the interac-
tions that generate them: most land-use activities affect only
a few environmental parameters, which, in turn, alter water-
shed processes and affect beneficial uses.
Our present knowledge of CWEs is based on many years
of research in several fields, and this research has taken a
variety of forms. Chapter 3 describes the strengths and
weaknesses of research approaches used in understanding
watersheds, ecosystems, and CWEs.
Chapter 4 discusses a variety of CWEs and categorizes
them according to the types of interactions generating them.
Studies that describe or predict CWEs are outlined, as are
existing methods for assessing land-use impacts. A program
for developing and validating a general CWE assessment
procedure is outlined. The proposed evaluation procedure
depends on a fundamental understanding of how CWEs
occur, and the remaining chapters outline the present state
of our knowledge.
The focus of this review is CWEs generated by use of
forest lands in California. Because CWEs include the effects
of any combination of activities occurring in a watershed,
methods for predicting CWEs must be able to assess the
effects of all land uses in a watershed and their relation to
naturally occurring processes. Interactions between impacts
related to timber management and those arising from other
uses must therefore be considered. Chapter 5 describes
research that evaluates the effects of various land-use activities
on environmental parameters, and Chapter 6 reviews studies
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
of how altered environmental parameters affect watershed
processes.
Resources and values influenced by CWEs are also numerous.
Of particular concern are impacts on fisheries resources,
timber use, recreation, water supply, and floodplain use, and
each of these resources has characteristic environmental
requirements. Chapter 7 outlines research that evaluates
environmental requirements and effects of altered conditions
on specific beneficial uses.
5
Chapter 2
How Cumulative Watershed Effects Occur
C WEs can result from the accumulation of effects
through time, through space, or both. They occur because
of fundamental properties of systems with interacting
components, and because of special qualities of watersheds
and ecosystems. To understand the basis for CWEs, it is
necessary to understand how watersheds and ecosystems work.
Watershed Function
A watershed is a topographical form that concentrates run-
off. The form is molded by transport of sediment or dis-
solved material from one part of the landscape to another,
generally with the aid of water.
Runoff Generation
Surface runoff from a watershed is produced by groundwa-
ter seepage and by water falling onto the land surface. When
water accumulates on a surface faster than it can infiltrate,
the excess runs off. Some infiltrated water may be conveyed
rapidly to channel networks through soil pipes, but the re-
mainder percolates through soil and bedrock until it encoun-
ters the water table and is incorporated into the groundwater
reservoir. Along the way, it may be absorbed by vegetation
and transpired back into the atmosphere, or adsorbed onto
soil particles and later evaporated. Springs and bogs form
where a water table intersects the ground surface, and seeps
can subtly augment base flow along a channel. Overland
flow contributes directly to runoff only during or immedi-
ately after snowmelt or rain; otherwise, streamflow is sup-
ported only by groundwater.
By dissolving chemical constituents of soil and bedrock,
percolating water speeds their physical and chemical
breakdown. This contributes to formation of transportable
sediments, aids weathering, and physically removes much of
the original bedrock mass. Limestone caves and sinkholes
are obvious evidence of chemical dissolution, but the process
also contributes to sculpting of most other landforms. Chemicals
produced by human activities can also be redistributed by
groundwater flow.
Channels
Water exerts force on surfaces it flows across, and this shear
stress increases with flow depth and gradient. High shear
stress enables runoff to mobilize sediment and excavate
channels. Elevated discharges transport more sediment than
lower discharges at a site, but the pattern is more compli-
cated along a channel’s length. Downstream increases in
6 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
flow depth usually compensate for characteristic decreases
in channel gradient. In an equilibrated stream, each channel
segment is precisely adjusted to carry off sediment contrib-
uted from upstream and from tributaries. Such a channel is
described as “graded.”
A channel’s cross-sectional form and long profile depend
on the balance between its ability to transport sediment and
the type and amount of sediment provided to it. Sediment
accumulates where input rates are higher than transport rates
or where contributed sediment is too large to move, and
deposition can locally increase transport by steepening a
channel or broadening its transport zone. Where flow can
carry more sediment than is provided, excess shear stress
acts on the bed and may cause incision. Down-cutting decreases
channel gradient locally but increases it immediately upstream,
increasing potential sediment transport into the reach while
decreasing transport out.
Channels adjust when conditions change, but conditions
change continually. Introduced sediment may accumulate
during low flows and be washed away when flows increase,
and an uprooted tree may cause local erosion by forcing
flow against a bank while allowing sediment to accumulate
in the slackwater it forms. Undisturbed channels continually
adjust to the distribution of high and low flows. Bars are
molded and channels realigned to accommodate flood
discharges, but subsequent low flows more weakly readjust
channels according to their own patterns of shear stress.
Channel form usually most strongly reflects moderate flows,
which are frequent enough to reassert their effects but large
enough to move sediment efficiently. Localized scour and
deposition may occur, and channels may migrate by bank
erosion and bar deposition, but the overall form of an
undisturbed channel generally changes little through time.
Fallen trees accumulate in streams until removed by floods,
debris flows, or decay. Woody debris traps sediment and
creates plunge pools and obstructions, which expend flow
energy that might otherwise contribute to erosion or sediment
transport. Sediment traps formed by debris can moderate the
impact of large inputs of coarse sediment by releasing stored
sediment gradually over decades or centuries.
Sediment Production
Several processes transport sediment to channels. On steep
slopes or in weak materials, gravity may nearly overcome
the resistance produced by inter-particle friction, cohesion,
and anchoring roots, and landslides can then be triggered by
increased pore-water pressures or loading by rain. Suscepti-
bility to sliding can change with soil age as weathering alters
cohesion, frictional properties, and soil depth. Shallow de-
bris avalanches commonly occur in bedrock hollows, where
pore-water pressures tend to be high and soils are deep.
Debris avalanches that land in steep channels may flow
downstream and incorporate channel sediment to form a
debris flow or debris torrent. Debris flows can continue to
move for several kilometers and often come to rest only on
entering lower gradient channels, where they form debris
jams of logs and sediment. Other types of landslides are less
influenced by topography. Clay-rich or sheared bedrock may
form earthflows over a kilometer long and hundreds of
meters wide. Earthflows move slowly during the dry season,
but they can move several meters in a month when saturated
by winter rains.
Gravity gradually pulls the entire soil mantle downslope
by a group of processes known as soil creep. Some soils may
flow plastically, much like a very slow earthflow. Others
move because of the downslope component that gravity
introduces to most soil disturbances. Soils expand
perpendicularly to the soil surface when wetted or frozen, so
they expand preferentially downslope. As they dry or melt,
upslope contraction against gravity does not fully compensate
for the original downslope motion. Similarly, when soil
particles fall into an animal burrow or empty root hole, they
move slightly downslope because they fall toward the center
of the earth rather than perpendicularly to the slope. Soil
creep can move a soil profile a fraction of a millimeter to
several centimeters each year.
Several transport processes primarily affect the soil surface.
Rainsplash launches particles preferentially downslope or loosens
them for entrainment by overland flow, and surface runoff
dislodges and transports sediment. Soils are protected from
both rainsplash and sheetwash erosion if they are covered by
vegetation, organic debris, or immobile particles.
Interactions between Hillslopes and
Channels
Sediment transported down a hillslope eventually encoun-
ters a stream channel. Entry of sediment into a channel
(“sediment production”) may result directly from hillslope
transport processes, as when a debris avalanche falls into a
channel. In contrast, some creep processes cause channel
banks to encroach gradually on a channel. As banks move
inward, channel flows deepen and increase the force exerted
on banks, and sediment production occurs by bank erosion.
Where stream channels are lined by floodplains, much of the
sediment removed from hillslopes accumulates on the flood-
plains until channel migration entrains it.
A channel aggrades if its ability to remove sediment
cannot keep pace with the amount introduced. Aggradation,
in turn, modifies sediment input rates: not only do sediment
deposits buttress hillslopes and reduce their transport rates,
but they also trap more hillslope sediment. In contrast, if
channels remove sediment more quickly than it is produced,
then they incise, hillslopes steepen, and hillslope erosion
rates increase to equal removal rates. Meanwhile, these
morphological changes affect the ability of channels to
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
transport sediment. Aggradation provokes changes that usually
increase channel transport rates, whereas incision may
decrease gradients downstream and thus reduce transport
rates out of a reach.
A watershed thus modifies its form by eroding or aggrading
to balance rates of sediment input from hillslopes against
rates of removal by runoff. This adjustment is realized by a
simple feedback mechanism: what does not get carried off
remains, and this alters both input and export rates. Aggradation
tends to increase transport capability and decrease input rate,
while incision increases the input rate and decreases transport.
The implications of this balance are extremely important for
land management. When one factor in a system changes, that
factor alters others to compensate for the change, and both
form and processes are likely to be modified in response.
Physical Basis for Cumulative
Watershed Effects
Most land use directly alters only a few environmental fac-
tors: vegetation, soils, topography, and chemicals. All other
changes result from alterations to these factors. CWEs are
caused by changes that accumulate in time, space, or both.
Accumulation of effects through time requires that indi-
vidual changes be persistent, so that one effect is not fully
healed when the next occurs. This mechanism can generate
on-site CWEs, but off-site impacts can be caused only if
changes accumulate through space, and thus off-site changes
must involve altered transport processes or watershed prod-
ucts. Land-use activities that cause persistent changes or
affect transport of watershed products thus have the poten-
tial for causing CWEs.
The importance of a cumulative change depends on what
resources or values are affected by the change. In many
cases the impact of concern is a physical one: accelerated
channel migration may destroy agricultural land, enhanced
peak discharges may increase flood frequencies, or increased
erosion may fill reservoirs. Off-site effects require alteration
of transport processes or rates, and these alterations often
affect channel morphology.
Ecosystem Function
Most land use alters biological communities, which can then
modify physical conditions by influencing production, qual-
ity, and transport of water and sediment. Watershed pro-
cesses, in turn, mold the physical habitat of biological com-
munities, and the resulting impacts on biological resources
are often a major concern. Biological processes affect both
the generation and expression of CWEs and must be under-
stood if CWEs are to be evaluated.
Because the network of interactions between organisms is
intricate, a biological change can affect an entire community
7
of organisms. Even those interested only in the abundance of a
particular species must take into account its community
interactions. A population change may result indirectly from
changes to another organism, and the resulting change may
affect still other organisms. To understand how biological
changes lead to CWEs, it is necessary to understand how
biological communities function, and this is the focus of ecology.
Krebs (1978) defined ecology as “the scientific study of
the interactions that determine the distribution and abundance
of organisms.” Organisms respond to their physical
environment, to the assemblage of species they interact with,
and to other members of their own species. Ecology is thus
concerned with extremely complicated interacting systems of
biological and physical influence. These influences are usually
evaluated at the scale of individuals, species, or communities.
Constraints on Individuals and
Species
Physical constraints provide a clear example of limits to
distribution of a species, because each species has physi-
ological requirements for temperature, moisture, light, and
environmental chemistry. However, within these broad lim-
its are a series of less-well-defined constraints. An individual’s
ability to reproduce, grow, compete, or survive disease is
often impaired near physical limits. The species may survive
under marginal conditions, but it may not thrive.
Physical constraints are complicated by other factors.
Physiological demands often vary seasonally or with
developmental stage, and the ability to compete under
particular physical conditions may depend on the identity of
competing species. Apparent physical constraints can also
reflect behavioral influences. Animals may frequent an
environment because it provides opportunities for nuptial
display, a vantage for spotting prey, or simply because of
historical happenstance. For example, dense seabird colonies
may occupy particular islands while adjacent ones remain
inexplicably barren.
Organisms interact with others and so are influenced by
their biological environment. Most animals are affected by
species dependent on them for nutrients, and in turn depend
on nutrients derived from other organisms. Both plants and
animals compete with others for resources and living space.
Habitat
An organism’s physical and biological environment is its
habitat. Sedentary species use habitats of fairly uniform
character, but migratory and wide-ranging species often have
a variety of habitat requirements. For example, some salmon
begin life in a stream, over-winter in a lake, and spend
several years at sea before returning inland to spawn. Some
organisms require different environments for different uses,
such as feeding and escape from predators. Welfare of a
species depends on the condition of each component of its
habitat. Characteristic and identifiable habitat components,
8 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
such as pools in a stream environment, are described as
“habitat units.”
For an environment to be habitable by an organism, it
must fall within the range of conditions the organism can
tolerate. This view of habitat has led to the concept of “limiting
factor”: abundance of an organism in a given environment is
assumed to be limited by a dominant constraint. Thus, smolt
production in some rivers is thought to be restricted by lack
of clean spawning gravels, while insufficient rearing habitat
limits production at other sites. If the limiting constraint is
removed, then abundance is expected to increase until it is
bounded by a different constraint. “Carrying capacity” is the
potential abundance of a species, given the physical and
biological environment at a site. Both of these concepts are
widely used in impact analysis, yet both are controversial.
A limiting factor is usually viewed as a single parameter,
although abundance is actually modified by interactions between
many physical and biological factors. Occasionally a single
factor may dominate, as when chemical spills or elevated
stream temperatures kill fish. However, abundance more
commonly reflects not only the local habitat, but also the
history of changes at a site, the history of the population, and
influences throughout the organism’s life cycle in every
component of its habitat. One factor may produce the highest
mortality one year, while the following year a different one
may dominate. In other cases, a factor may become important
only because an interacting factor changes. For example,
elevated turbidity or toxins might harm fish only while water
temperature is high. The concept of limiting factors can be
applied either to a population, in which case all influences
throughout the population’s life cycle must be considered, or
to an organism’s tenure in a particular habitat unit, where only
local influences are of concern.
“Carrying capacity” also can be applied in different
contexts. Carrying capacities have been defined for particular
habitat units during specific seasons, but have also been
measured as time-averaged values in systems containing
many habitat units. Carrying capacity for a system must
account for the range of habitat conditions present through
time; it cannot be defined simply by characteristics observed
at a particular time. System capacity must also reflect the
full range of environments used by the species. Whether
carrying capacity is defined for a system or a habitat unit, it
is usually poorly correlated with abundance. Abundance
also depends on nonhabitat factors such as disease, abundance
of the previous generation, and recent history of disturbance.
Biological Communities
Some biological environments are recognizable entities: in
California, a redwood forest in Del Norte County looks like
a redwood forest in Humboldt County, and is very different
from a red fir forest in the Sierra Nevada. Once the forest
type is identified, its plant and animal species can often be
predicted. These environments are biological communities.
A community is usually described using the types and rela-
tive abundance of species comprising it and the structure of
the community. Communities have a physical structure im-
parted by the size and forms of plants, and a trophic structure
defined by who eats what or whom. The physical character-
istics of a site strongly influence which biological commu-
nity develops there.
Biological communities are opportunistic collections of
individuals brought together by overlapping sets of physical
and biological requirements, and how a community functions
depends on how its organisms interact. Each organism is
part of the environment of others, and each has a particular
role as a consumer and provider of resources. These roles
must be filled if the community is to remain stable. Functional
roles within a community are called “niches,” and organisms
with overlapping niches compete with each other. If a habitat
change depletes a species, competitors better able to tolerate
the new conditions may benefit from its absence and multiply
to fill the niche. Removal of a habitat unit or species can also
destroy a niche, as when a prey species is replaced by an
unpalatable competitor.
Community members thus participate in a complex network
of interdependencies. If abundance of a species changes, it
will usually cause adjustment throughout the network. When
an accustomed predator is removed, prey populations may
increase to the point that food resources are depleted and
their competitors impacted, while a reduction in the population
of a prey species may decrease the abundance of its predators.
Response to Change
Biological communities undergo continual change. Organ-
isms and species respond to changing environmental condi-
tions on scales of minutes to millennia, and as they vary in
abundance, so do organisms dependent on them. Until hu-
mans arrived in the Americas, large-scale community changes
in the New World generally resulted from climatic shifts.
These shifts provoked slow migrations of plant and animal
species and changed both the composition and distribution
of most communities.
In contrast, short-term and localized disturbances have
more restricted effects. Forest fires, windstorms, and floods
alter the physical structure of communities and allow
colonization by successional species, but the communities
that are eventually reestablished are usually similar to the
original. Successional assemblages can even be viewed as
parts of the community, just as forests include species that
colonize windthrow gaps. Physical disturbance is often
essential for maintaining the character of a community and
preserving species diversity.
On an even shorter scale, organism requirements and
interactions vary by season. Food sources and predation risk
change regularly through the year, and behavioral patterns
have developed to take advantage of predictable habitat changes
in stable systems. For example, wildebeest migrate to Tanzania’s
Serengeti Plain just as new grass is ready to support them.
The response of a biological community to changing
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
conditions depends on the nature of the trigger and the types
of organisms and interactions in the community. Some
communities can shift through time in response to gradually
changing conditions without experiencing major disruptions.
“Resilient” communities may change radically to adjust to
short-term changes, yet can later reestablish themselves in
their original form. Others, described as “stable” or “resistant,”
can absorb major changes in conditions without being altered.
Still others respond with a complete change in character.
Most communities are fairly stable or resilient under
natural conditions. Catastrophes like floods or fires may
occasionally destroy segments of a community, but its
biological and physical context is unchanged so the regenerated
community is much like the original. However, if the biological
or physical context is altered by land use, or if the frequency
of disturbance changes, then populations will adjust to reflect
the new conditions. As populations change and provoke
compensating changes in interacting populations, the nature
of their communities also changes. Biological change is thus
measured not in terms of a single species’ abundance, but by
characteristics of its community.
Sensitive “index species” are often monitored to detect
changes in communities for impact analysis. The example of
coal-mine canaries is frequently used to illustrate the concept:
toxic fumes killed the canary before they affected people,
allowing timely evacuation. However, survival of an organism
depends on many physical and biological factors. The canary
was a successful index species for air quality because its
reactions were understood and its limitations recognized;
canaries would have been unsuitable for warning of flooding
or collapse. Similarly, index species cannot provide warning
of all potential impacts, because some species are more
sensitive to some changes than others are.
Like watersheds, biological communities are intricate systems
of interdependent components, and when one component is
altered, those interacting with it must change in response.
Biological Basis for Cumulative
Watershed Effects
Land use usually alters vegetation. If this change is persis-
tent or affects transport processes or rates, it is capable of
generating CWEs. Persistence of a vegetation change de-
pends on the lifespan and age of the species affected, the
successional sequence, the type and persistence of impacts
generating the change, and the response mode of the vegeta-
tion community. In general, altered distributions of long-
lived species persist longer than those of short-lived ones,
and thus carry a greater risk of contributing to on-site CWEs.
Vegetation influences the transport of watershed products.
Hillslope and riparian vegetation affects generation of runoff,
sediment, and organic debris. As these watershed inputs
change, channel morphology adjusts to accommodate them,
and change propagates along channels as aggradation, incision,
bank erosion, changes in bed material, or altered channel
pattern. These changes, in turn, affect water velocity, flow
9
depth, baseflow discharge, intergravel flow, water temperature, National Park threatened the stability of Tall Trees Grove,
water chemistry, and flood frequency. and a desire to control this impact was an important motivation
The habitats of all aquatic and riparian organisms are for expansion of the park.
affected when channel morphology changes. Alterations may Reid and Miller (1989) discussed the causes for
occur within habitat units (e.g., changes in temperature or documented species extinctions, and noted that most
flood frequency), in their distribution (e.g., fewer undercuts), ecosystem disturbances and extinctions are the result of
or in their variety (e.g., infilling of pools). As the habitat cumulative effects: “Any factor that leads to a decline in the
changes, so do the populations of organisms dependent on it. population size of a species makes it more vulnerable to
Any change in species abundance affects populations of the extinction” (Reid and Miller 1989, p. 45). They identified
organisms it feeds on and those that feed on it. In this way, habitat loss, degradation, and fragmentation as the most
altered channel morphology can propagate changes throughout important factors in extinctions, followed by overexploitation,
biological communities. species introductions, and pollution.
Impacts on biological communities can be cumulative in
several ways. Different land-use activities can all contribute
to the same impact, as when urbanization, road construction,
agriculture, log-milling, and other activities each A Framework for Understanding
independently destroy off-channel rearing habitat for fish.
Although each activity may be responsible for only a 10 CWEs
percent decrease in rearing area, a combination of ten such
activities would result in its complete loss.
Many on-site and downstream interests may be affected by a
Different activities can also produce complementary
single land-use activity, and many activities may occur within
impacts. Gravel mining may decrease spawning success by
a watershed (fig. 2). If the effect of each potential combina-
destabilizing channels at the same time that timber management
tion of activities on each potential downstream resource had
activities decrease over-winter survival and fishing decreases
to be measured, prediction of CWEs would be intractable.
adult populations. Each activity affects the fish population in
Figure 2, for example, includes more than 50,000 combina-
different ways, but the cumulative effect is decreased abundance.
tions of multiple land uses and impacted interests. The prob-
Repeated impacts at the same site can cause a cumulative
lem can be greatly simplified by focusing on the relatively
effect. A species’ mechanisms for coping with impacts are
few mechanisms by which impacts occur.
often keyed to the characteristic frequency of disruption
All impacting activities directly affect only a few
with which it has evolved. If the disturbance frequency is
environmental parameters, including chemical content,
increased (as by logging) or decreased (as by fire control),
vegetation, topography, soil properties, and fauna. Different
the community will change in response. Anadromous salmon
activities often affect environmental parameters in the same
cope with occasional catastrophes in part by having some
way, and changes can be measured for each activity. The
variation in the age of returning spawners. If a year class is
persistence of each type of change and its effects on the transport
destroyed by excessive scour during a storm, a few returns
from other age classes ensure that the year class is eventually
grazing water use agriculture
reestablished. However, if sequential age classes are destroyed LAND-USE
mining recreation urbanization
ACTIVITIES
by repeated catastrophes, as might occur if land use increases roads fishing flood control
power navigation logging
scour frequency, the species’ coping mechanism is defeated, etc.
and the population may collapse. chemicals soil ON-SITE
ENVIRONMENTAL
Cumulative impacts can also occur when one activity makes PARAMETERS
vegetation water BENEFICIAL
topography fauna USES
a population vulnerable to a different type of change. Loss of
tributary habitat might force young fish to over-winter in logging
esthetics
mainstem channels at the same time that hydrologic alterations wildlife
TRANSPORT OF water sediment habitation
are increasing flood frequency and decreasing in-channel survival. WATERSHED organic chemicals grazing
PRODUCTS material heat
Several types of biological impact have attracted notice recreation
agriculture
in the past. The most frequent trigger for popular concern is etc.
decreased abundance of a desired species. Decreased duck
power flood control agriculture
and geese populations provoked organization of Ducks OFF-SITE grazing esthetics water use
BENEFICIAL fishing habitation transporta-
Unlimited, which established refuges to offset the cumulative USES logging recreation tion
impacts of progressive wetland conversion. Changes can wildlife etc.

also affect the characteristics of a species, as where high

fishing rates and large hatchery releases have combined to
decrease the genetic diversity of some anadromous fish
species. In other cases, concern arises from a site-specific
alteration. Aggradation of Redwood Creek in Redwood
Figure 2—Framework for understanding cumulative watershed ef-
fects. Land-use activities affect a small number of environmental
parameters, and these can alter production and transport of water-
shed products. Off-site impacts can result only from a change in
transport. Arrows indicate an influence of one category on another.

10 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

of watershed products (water, organic material, sediment,
chemicals, and heat) can then be evaluated independently of
the triggering land uses. Other resources can be impacted only
by changes in environmental parameters or by altered transport,
and these impacts can also be defined independently of triggering
land use. Off-site CWEs must result from changes in transport,
because impacts can be propagated from one location to another
only if something is moved there. At each stage, environmental
parameters and transport mechanisms can interact to alter the
expression of impacts.
The problem of understanding CWEs thus becomes: (1)
understanding the effects of specific land-use activities on
physical and biological environmental parameters, (2)
quantifying the influence of altered environmental parameters
on watershed products and transport mechanisms, and (3)
understanding how changes in environmental characteristics
and transport mechanisms affect particular resources and values.
CWEs can be caused by repeated, sequential, or coexisting
land-use activities, or by activities that progress either through
time or across a landscape. Activities may produce the same
type of change to environmental parameters, cause different
changes that contribute to the same impact, influence another
activity to cause an impact, or provoke changes that interact
with each other to produce an impact. These mechanisms
will be discussed in Chapter 4.
Issues that Complicate the
Understanding of CWEs
Physical and biological systems that generate and experi-
ence CWEs are extremely complex and interact intimately
with each other. This complexity creates a variety of prob-
lems for CWE analysis.
Off-site CWEs result from altered transport of watershed
products. Because transport usually involves redeposition and
re-entrainment, watershed response usually lags behind the
changes driving it. Time lags between cause and effect can
obscure the reasons for environmental change. In some cases,
ongoing changes result from activities carried out centuries
ago. In the Appalachian Piedmont, for example, today’s high
sediment loads include sediments originally loosed by
exploitive agricultural practices of the 1700s (Costa 1975).
Also because transport is involved, impacts may occur far
from the activities triggering them. Gilbert (1917) recognized
this in his study of effects of hydraulic mining in the Sierra
Nevada on navigation in San Francisco Bay, 200 km
downstream. This geographic decoupling also masks reasons
for change and causes time lags in impact expression.
Causes of change are also obscured because different
sites respond to a particular environmental change in different
ways. Low-gradient reaches aggrade more readily than steep
reaches, for example, so coarse landslide sediment may have
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
little effect until it encounters a susceptible reach. Such sites
are common where mountain channel gradients decrease on
entering intermountain valleys. Lyons and Beschta (1983)
associated historical changes in channel pattern of Oregon’s
Upper Middle Fork Willamette River at such a site with
forestry-related landslides high in the basin.
The progress of change can be recognized as it occurs if a
system responds incrementally, as when reservoir siltation
rates increase with increasing watershed disturbance.
However, responses such as landsliding and gully formation
involve discrete changes in state and are not evident until
they occur. Discrete changes are often presumed to be triggered
when a threshold value is exceeded, but they usually represent
the combined effects of multiple influences. For example, a
slide might occur because bedrock has weathered to the
point that the amount of water added to the soil during a
storm, with the addition of culvert outfall, can permit the
accompanying wind to topple a tree and trigger the failure. A
threshold has been surpassed in that a response has been
provoked, but the threshold cannot be identified as a discrete
value of a single parameter.
Apparently benign changes may accumulate that require
a trigger of sufficient magnitude to exhibit their full impact.
Culverts designed to carry 25- or 50-year flows are common
in the western United States, for example, but the cumulative
impact of this practice will not be apparent until a larger
storm occurs. Slopes destabilized by land use can show a
similar response: nothing may happen until a large storm
occurs, and then sliding may be widespread. This response
confuses interpretation of the role of management in generating
impacts. The storm of December 1964 was the largest on
record in parts of California. Landsliding and channel
disruption were widespread, but how much would have
occurred before land-use disturbance? Analysis is difficult
without precedents or undisturbed catchments for comparison.
In other cases, chronic changes may take a long time to
accumulate to the point that adverse effects become evident.
Siltation of a large reservoir may not affect us, but it may
economically cripple a region 100 years from now. Similarly,
global warming began with the industrial revolution, but its
potentially devastating effects are not yet conclusively measurable.
Different changes interact and modify a system’s response,
further complicating CWE analysis. Thus, toxicity of an
introduced chemical might increase with acidification of the
stream system (Cleveland and others 1986), or decrease with
addition of suspended sediment (Hall and others 1986).
Finally, principles of equifinality and indeterminacy
undermine the confidence of prediction and understanding.
Equifinality refers to the concept that different causes may
result in similar responses. Thus, the cause of an impact
cannot be diagnosed simply from its nature, just as a virus
cannot be identified by the fever it produces. Indeterminacy
expresses the opposite idea: a particular change may not always
elicit the same response. The expression of an environmental
change reflects many modifying conditions, and those conditions
must be understood if a response is to be predicted.
11
Chapter 3
Research Methods for Understanding Cumulative
Watershed Effects
C umulative watershed effects will no longer be an
opera-tional issue when the combined effects of land-
use activities can be anticipated, causes of particular impacts
diagnosed, and effects of mitigation or prevention measures
predicted. Progress toward resolution of the CWE problem
is measured by how well these tasks can be carried out, and
CWE research includes any research that contributes to these
goals. Three phases of research are required:
1. Provide a basic understanding of how and why CWEs
occur.
2. Construct a method for evaluating CWEs.
3. Test the effectiveness of evaluation methods and
mitigation programs.
The most challenging phase, and that on which the others
depend, is to develop an understanding of how CWEs occur,
and thus of how physical and biological systems work. Most
research carried out in the fields of geomorphology, hydrology,
and ecology contributes to our understanding of CWEs, but
very little research has specifically focused on CWEs. This
chapter discusses the strengths and weaknesses of common
research approaches as they are applied to understanding
CWEs, and is intended to provide a basis for evaluating
studies described in later chapters.
Requirements of CWE Research
The CWE perspective introduces three issues of special
importance: large spatial and temporal scales, complexly
interacting systems, and an interdisciplinary focus. Each of
these is considered a difficult problem, and past studies were
often designed to avoid them. Studies that have dealt explic-
itly with these issues are particularly relevant to CWEs.
CWEs involve the accumulation of impacts over time or
through space, and these time or space scales may be large.
Research must account for long-term variation in conditions
and processes, and for variations over wide areas. The effects
of infrequent events, such as large storms or fires, need to be
understood, as well as the influence of the spatial distribution
of activities and impacts. Studies often use data collected at
a variety of scales, but methods for reconciling multiple
scales are poorly developed.
Many CWEs are caused by interactions between different
processes. Such interactions have often been excluded from
studies as secondary complications, or are overlooked because
12
they fall between disciplines. Most experiments are designed
to simplify systems by avoiding interactions.
Evaluation of interactions is even more difficult when
they are between physical and biological systems, as is
usually the case with CWEs. These evaluations require
interdisciplinary work, but this has also often been avoided.
Multidisciplinary work, where researchers from different
disciplines work at the same site, is occasionally substituted
for interdisciplinary studies. Multidisciplinary work often
provides a broad perspective on conditions at a site, but
without close understanding between disciplines and a research
plan that specifically addresses interdisciplinary problems,
it is no guarantee of interdisciplinary understanding.
Research Approaches and
Methods
Some research approaches are more productive than others
for addressing particular aspects of CWEs. Which approach
is selected depends on the nature of the problem, how it is
posed, the type of solution desired, and the level of funding.
A system can be viewed as a collection of parts, where the
function and response of component parts explains the system
response. Systems can also be represented by a series of processes
that are activated to produce a response. Alternatively, systems
can be viewed as black boxes: certain stimuli provoke a given
response, but the mechanisms producing that response are
irrelevant. A system can be studied from any of these points of
view by selecting an appropriate research approach. A single
investigation often uses several approaches, as when a study of
a watershed process includes monitoring, manipulation, and
modeling phases. In other cases, studies using one approach
require data produced using others. Each research approach
has inherent strengths and weaknesses (table 1).
Descriptive Studies
The early phase of most sciences is descriptive and often
makes use of case studies, in which particular incidents or
sites are examined in detail. Phenomena are cataloged, mea-
sured, and described to disclose patterns of occurrence. De-
scriptive studies might address questions such as: “In what
ways did this stream change when land use in the basin
changed?” or “How much did salmon populations change
when this dam was constructed?”
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Table 1—Advantages and disadvantages of research approaches and methods
Approach or
method Applications and advantages
Descriptive Useful for exploring a new problem
study Can identify interactions
Helps define research questions
Allows detailed observations
Requires little prior knowledge
Extensive Often used for inventories or to develop
survey databases for statistical analysis
Reveals patterns of occurrence
Defines spatial distribution of effect or
process
Identifies associated variables
Experiment Defines cause-effect relations
Variables can be isolated to make a
problem tractable
Identifies controlling variables
Produces generalizable results
Can be inexpensive and quick
Modeling Can predict results for hypothetical
scenarios
Reveals data requirements
Shows sensitivity to variables
Helps organize information
Reveals gaps in understanding
Monitoring Reveals change
Defines baseline conditions
Provides data for comparison
Reveals temporal association
Implementation is methodical
Statistics Useful for describing systems
Reveals patterns of association
Measures strength of association
Distinguishes effects of multiple
variables
Can use to calculate probability
Descriptive studies identify processes and interactions active
at a site, often with the expectation that they will be found
elsewhere in similar situations. These studies frequently identify
questions to be addressed later using other techniques.
Descriptive studies have the disadvantage that numerical results
cannot be applied to other sites unless patterns of variation
are understood, and how widely applicable even qualitative
results are is unknown without further survey work.
Dietrich and Dunne’s 1978 study of sediment production
in a forested watershed is a descriptive case study. Although
numerical results are relevant only to that case, the patterns
of process interactions that were identified, the approaches
taken in analysis, and the framing of the problem all provide
useful guidance for other studies.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Disadvantages
Results cannot be generalized to other
sites or times
Nonroutine methods are often required,
so personnel must be knowledgeable
Requires large sample sets
Often expensive unless data already exist
Can show association, but not cause
Results cannot be generalized beyond
sampled area or time
Implementation is usually methodical
Design of sampling plans requires prior
knowledge
Need prior understanding to design an
effective study
Results using isolated variables may be
misleading if variables interact
Requires sophisticated knowledge of the
system to be modeled
Complexity makes model opaque
Calibration may hide defects in model
Often requires a lot of data
Results of complex models are hard to
validate
Long-term sampling is often required
Can show association, not cause
Cannot generalize to other sites without
other information
Must be carefully planned to provide
valid results
Requires long-term budget commitment
Often misapplied: wrong methods are
used; or method is applied to cases that
violate assumptions; or results are
extrapolated beyond sampled range
Extensive Surveys
Extensive surveys measure parameters or make qualitative
observations at many sites to detect patterns of occurrence.
These are commonly used to map distributions of processes
or attributes, determine their significance, or reveal factors
associated with them. Extensive surveys include both
inventories, such as soil or vegetation surveys, and sampling
for statistical analysis. These techniques can be used either
to describe a phenomenon or to associate occurrences with
particular conditions. The approach might be used to address
such questions as: “How much sediment do debris flows
contribute to streams in this area?” or “What soil,
topographical, and land-use variables might explain the
distribution of landsliding in the region?” Extensive surveys
can be used to identify either the responses associated with a
13
particular environmental change or the environmental changes
associated with a particular type of response.
Inventories and statistical surveys are fundamentally
different. Inventories sample an entire population and so
produce databases that fully represent the population. A soil
map, for example, provides information at every point within
the sampled area. Computer-based Geographic Information
Systems (GIS) are useful for managing inventory information.
In contrast, statistical surveys sample a subset of a
population and use the information to infer the full
population’s characteristics. Precise spatial information is
usually not needed, although results are often used to infer
distributions. For example, a statistical survey might show
landslides to be most common on shale slopes steeper than
30°. Bedrock and slope maps could then be used to map
landslide potential in the sampled area.
Samples must represent the range of spatial variation likely
to occur in a study area. Sample selection may be completely
random, or random within identified strata, as when 20 samples
are randomly selected on each of three rock types. Stratified
random sampling usually requires fewer samples to produce
the same precision as completely random sampling, because
sample variance is reduced by segregating samples according
to meaningful strata. The potential efficiency of a sampling
plan thus increases with understanding of a feature’s controlling
variables. Sampling is often carried out at fixed intervals, as
when soil depths are measured at intervals along road cuts,
but interval sampling may produce invalid results if the
measured property is not randomly distributed.
Many statistical methods are available for disclosing
patterns among survey data. Multivariate techniques are
particularly useful for exploratory studies, which often
measure many variables to identify those most closely
associated with a phenomenon. Discriminant analysis
identifies the combination of variables best able to predict
membership in selected categories. Cluster analysis defines
groups having similar characteristics, and factor analysis
identifies associations between variables.
Extensive surveys are often used to identify patterns,
associations, or distributions for use in predicting response
to change. Exploratory studies frequently use the method to
define important issues or to reveal possible controlling
variables. However, statistical surveys merely identify
association; they cannot demonstrate causality.
It is tempting to generalize results of extensive surveys to
wider areas or longer time periods, but generalization is
tenuous unless the reasons for statistical association between
variables are known. Valid generalization also requires that
a wide enough variety of conditions be sampled that variations
in driving variables average out. Qualitative results are more
readily generalized than quantitative ones, but even they
cannot be applied to areas with different site characteristics
or to the sampled area if conditions change.
Extensive surveys have several drawbacks. Many samples
must be measured if surveys are to provide valid statistical
results. If controlling variables are not already identified, the
14
sample size must be enlarged further to test the range of
potential controls. In many cases, a system’s prior history is
an important control, but this can be difficult to assess. For
example, effects of the 1964 storm lasted for at least a
decade in many west-coast watersheds, and surveys carried
out during the recovery period reflect the storm’s influence.
Local events also introduce variation in system response and
further increase the sample sizes required to distinguish
results from noise.
Extensive surveys also have several strong advantages.
They can be used to identify possible controlling variables
and assess spatial distributions, and survey implementation
is relatively routine once sampling protocols are defined.
Anderson (1954) used an extensive survey to determine
the effects of several land-use activities and environmental
parameters on sediment yields. Multiple regression showed
which land uses are associated with increased yields, and
allowed mapping of erosion potential in western Oregon.
More recently, Touysinhthiphonexay and Gardner (1984)
measured channel geometry in 29 basins sustaining different
intensities of strip mining and demonstrated an association
between proportion of area mined and the magnitude of a
channel’s response. Lewis and Rice (1989) measured attributes
of 655 logged units in California and used discriminant
analysis to identify parameters associated with erosion events.
In each case, the study’s goals were to provide generalizable
information for use in predicting watershed response to land
use and to identify factors that control the response.
Experimentation
Experiments are used to identify and quantify relations be-
tween cause and effect. Experimentalists control the context
of a system’s response by holding all variables constant
except those being tested. Test variables are then manipu-
lated, and responses are attributed to their influence. Experi-
ments usually are interpreted by comparing results from an
untreated sample or from the same sample before treatment.
Questions addressed by experiment might include: “How
does clearcutting affect baseflow?” and “What pore pres-
sures will induce landsliding?”
Experimentation is more effective than monitoring or
surveys for revealing cause-effect relations, but success
depends on how well variation can be restricted to specific
variables. If other parameters vary, results become less
interpretable. For example, watershed-scale experiments often
test the effects of land use on basin response by comparing
either pre- and post-treatment measurements or measurements
in similar treated and untreated basins. In the first case,
differences in the character of storms during the two study
periods introduce magnitude-dependent and carry-over effects.
These effects are avoided if coexisting manipulated and
untreated basins are compared, but the basins will not be
physically and biologically identical. Small differences in
basin character can provoke different responses, and the
chance occurrence of an infrequent event in one basin (e.g.,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
a wildfire or landslide) will confuse the interpretation of
observed differences. Grant and Wolff (1991) used a long-
term monitoring record from a paired-basin experiment in
the Oregon Cascades to illustrate these effects.
Well-designed experiments can be used to isolate and
study parts of a complex system in ways that would be
impossible in an unmanipulated system. In addition, it often
takes less time and money to use a battery of experiments to
approach a general problem than to employ extensive surveys
or long-term monitoring.
Design of effective experiments requires knowledge of
the system being tested. Many parameters might influence a
system’s response, and an experimentalist either needs to
know which are the more important or must be willing to test
the lot. In addition, the researcher should know how controlling
variables interact, because isolation of a variable from a
closely correlated one may cause atypical responses. Response
of a complex system is usually determined by many linked
changes, so it is often difficult to isolate a particular aspect for
study. Some problems may require one set of experiments to
identify interactions among changing variables and a second
set to test response of the system to changes. Experimental
work is rarely routine, and participation by experts is often
required at each step.
Many CWE studies include laboratory and field
experiments to isolate interactions or processes for study.
For example, Hall and others (1986) and Cleveland and
others (1986) explored the effects of zinc and low pH,
respectively, on aquatic organisms and tested the influence
of other variables on those effects. Work was carried out in
laboratories to control for habitat variables. At a larger scale,
Collins (1987) controlled grazing and burning on field plots
to examine the effect of interactions of these disturbances on
prairie plant communities. In the field of geomorphology,
Bradford and Piest (1977) irrigated a gully wall to measure
pore pressures necessary for failure, and Wolman and Brush
(1961) used a laboratory flume to explore the effects of
several controlling variables on channel form in sands. Many
watershed manipulation studies have been carried out to
quantify various effects of land use (e.g., Harr and McCorison
1979, Rice and others 1979).
Modeling
Modeling uses physical or mathematical analogs to simulate
systems and to explore relationships between controlling
variables and system responses. Physical modeling has been
used to solve problems for millennia, and mathematical
modeling is becoming increasingly common with the spread
of personal computers. Problems addressed by modeling are
similar to those approached by experimentation, and in-
clude: “How would a change in baselevel affect gully ero-
sion?” and “How would a 30 percent increase in ground
cover affect runoff hydrographs?”
Physical modeling is often used as a tool in experiments,
as when studies use flumes as analogs for channels. If physical
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
models are of a different scale than the features modeled, they
must be carefully designed to ensure that changes in scale do
not introduce aberrant responses. A small-scale model of a
stream, for example, might require use of a fluid other than
water if sediment transport measurements are to be representative.
Of the approaches considered, mathematical modeling
requires the most sophisticated understanding of a system if
it is to be used successfully. In the ideal case, driving variables
are fully understood and are described mathematically, and
these relations are combined to predict system response to a
stimulus. Predicted results are often compared to measured
results to test a model’s validity.
Mathematical modeling is easy to misuse. Modeling results
are only as good as the data used to construct the model, and
data are frequently inadequate. Data needs are usually large,
and may include parameters that are rarely measured. These
requirements tempt non-field-oriented modelers to estimate
values for missing data. Results obtained using estimated
parameters may still be useful if estimates are supported by
field measurements from other sites, and if the sensitivity of
results to inaccuracies is determined. If several values are
estimated, sensitivity analyses must be carried out for
combinations of parameter values.
Missing data can also be managed by constructing models
that require calibration. However, calibration can hide flaws
in deterministic parts of a model, and simulations become
little more than regression exercises unless estimated variables
are identified and their effects understood.
Model complexity has grown with access to sophisticated
computers. The probability of error increases with complexity,
and errors become increasingly difficult to detect. Even
commercial software generally contains errors discovered
only after long use, and the frequency of errors in
noncommercial, unreviewed software is higher still. Because
errors are usually identified from anomalous results, those
contributing to results that fit users’ preconceptions are
particularly hard to recognize. A model’s validity also depends
on the validity of the assumptions it incorporates. Rarely are
these described in enough detail for other users to evaluate
them fully, and implicit assumptions may be hidden among
thousands of lines of computer code.
Proper model validation requires that predictions be
compared to measured responses throughout a model’s
intended range of use. If a model must be calibrated, then
data sets used for validation and calibration must be
independent of one another. Some models are intended to
predict effects of hypothetical changes, so validation over the
range of intended applications is not possible. Fully
deterministic models are most acceptable under these conditions,
but even these results are suspect.
Valid models are useful for predicting responses to
particular stimuli, as long as the inherent uncertainty of the
results is well understood. Modeling is often the only way to
predict response to unprecedented conditions.
Although a model’s predictive power depends largely on
its validity, modeling has other applications that do not. The
15
attempt to construct a model often produces its most useful
results: linkages between processes must be identified and
understood, information organized, important parameters
identified, and data needs clarified. Simulations can test a
system’s sensitivity to particular variables, further identifying
critical linkages and factors. Modeling also provides a measure
of how well a system is understood. Because anomalous
results reveal gaps in understanding, even unsuccessful models
can provoke useful questions and identify directions for
further study.
Predictive equations are also a type of model, and are
widely used to estimate flood flows (e.g., Muskingum routing
method, McCarthy 1940) and sediment production (e.g.,
Universal Soil Loss Equation, Wischmeier and Smith 1978).
In these cases, user groups are large enough that the models’
limitations are well defined (e.g., Wischmeier 1976), and
results have proved extremely useful. Hirschi and Barfield
(1988a) combined several physical relations into a model to
predict rill erosion. These authors explicitly presented their
model as a research tool, and, in a companion paper (1988b),
carefully described validation of the model.
Some simple models are distributed for their heuristic
value, and these may be useful tools for generating hypotheses.
For example, Ahnert (1976) described a 3-dimensional
landscape evolution model that can be used to explore the
effects of climate change on landforms. More complex models
are often used primarily by those constructing them, both
because others are hesitant to use products that may not have
been fully tested and because many models address problems
with stringent constraints.
Monitoring
Rather than being a research approach, monitoring is a method
that can be applied to many research approaches. Monitor-
ing studies measure attributes through time and are often
used to define site characteristics such as rainfall, distribu-
tion of flood peaks, and average sediment yield. The success
of future research depends on continued gathering of long-
term records, both to define baseline conditions and to detect
and measure responses to change. Many studies require
short-term monitoring. Time-sequence measurements de-
fine the response to an experimental manipulation, for ex-
ample, and modeling results are often tested against moni-
toring records. Monitoring provides answers to questions
such as: “Is global warming occurring?” and “How often is
this site likely to experience flooding?”
Like the spatially distributed samples used in extensive
surveys, monitoring samples may be selected according to
identified strata (e.g., storms and nonstorms), at fixed intervals,
randomly, or randomly at frequencies weighted by subclass
importance (as when turbidity samples are collected at a
frequency weighted by discharge). Most monitoring has used
fixed sampling intervals. These intervals determine resolution
(e.g., daily precipitation is of little use for predicting sheetwash
generation), and can introduce spurious results if the properties
16
being measured are periodic. A daily 6:00 A.M. stream
turbidity measurement, for example, is likely to underestimate
the effects of truck traffic.
To be valid, a monitoring plan must represent the range of
conditions present, provide enough measurements that short-
term fluctuations are averaged out, and ensure uniformity of
sampling technique. The particular research application
defines the precision, accuracy, and sample size required.
Data are usually analyzed statistically to identify trends,
describe the variance, and estimate probabilities and recurrence
intervals for events of particular magnitudes. Time series or
other sequential analyses are often appropriate but require
an understanding of possible interactions between sampling
and sampled periodicities.
Monitoring may quantify a site’s response to a change,
but response of an adjacent site cannot be predicted unless
the mechanisms for response are known. Monitoring produces
only circumstantial evidence of causality, because it can
demonstrate only temporal association between events. Many
of the existing short-term records of process rates are unusable
because their relation to long-term seasonal or annual variations
is unknown. Large data sets are usually necessary for adequate
sampling of the temporal variability at a site, and good baseline
data require long-term commitments of time and funds.
Despite its drawbacks, monitoring is often the most
effective method for identifying baseline conditions and
measuring changes through time, and short-term monitoring
is essential for quantifying experimental results. Once a
monitoring system is in place, sampling is usually routine.
Monitoring is an important part of many studies. Anderson
(1954) based his extensive statistical survey on monitoring
records, and Lyons and Beschta (1983) used existing stream-
gauge records to associate changes in peak flow with changes
in land use in the Upper Middle Fork Willamette River
basin. Emmett (1974) used a network of monitored cross-
sections to describe morphological changes in arroyos of the
southwestern United States. Megahan and Kidd (1972a)
installed catch troughs to monitor sediment eroded from
logged sites in Idaho.
Most experimental watersheds provide both long-term
baseline data and shorter-term records to document results of
watershed manipulations. Rice and others (1979) and Harr
and McCorison (1979) used both of these types of data to
assess hydrologic response to altered land use in experimental
watersheds in California and Oregon, respectively. The two
functions for experimental watersheds are not always
compatible, however, because basins must remain undisturbed
if they are to provide continuous baseline records. Thomas
(1990) described problems arising from use of experimentally
manipulated basins as controls in later experiments.
Statistics
Statistics, like monitoring, is more of a tool than a research
approach, and most studies use statistical analysis to identify
and quantify relationships between variables and responses
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
and to measure the strength of those relationships. Statistical
methods are required to answer such questions as: “How
often is this event likely to occur?” “Is this response atypi-
cal?” and “Which variable is most influential in producing
this result?”
No matter how sophisticated, statistics only describe
information already contained in a data set, so their usefulness
depends on data quality and study design. Enough samples
must be measured that treatment effects can be distinguished
from system noise, and enough variables must be controlled
that interpretation of the result is possible.
Because computers have made statistical methods easy to
use, statistics have also become easy to abuse. Computers
can lead naive users through complicated analyses without
addressing the methods’ applications, assumptions, and
limitations. As a result, inappropriate techniques are often
used, assumptions violated, and outcomes predicted beyond
the original data range. Sediment rating curves are extrapolated
to predict sediment loads during major floods, for example,
and recurrence intervals are estimated for unprecedented
storms. Mark and Church (1977) reviewed use of linear
regression in earth sciences, and found that in only two of 24
studies was the method applied appropriately. Benson (1965)
and Williams (1983) reviewed other types of statistical errors
common in hydrology and earth sciences. Some statistical
packages for personal computers contain undocumented errors
or faulty algorithms, and calculated results may not be valid
(Dallal 1988).
However, if the methods to be used are thoroughly
understood by the researcher, statistics are invaluable for
making sense of research results. Statistical methods allow
identification of patterns of response, calculations of
probability or risk of a particular occurrence, and evaluation
of the strength of relationships.
Scale and Focus of Studies
In designing a study, researchers select not only the ap-
proaches to be used but also the organizational strategy.
Strategies differ primarily in how problems are framed and
in the spatial and temporal scales considered.
Process Studies
Complex systems are often studied by isolating individual
physical or biological processes and measuring factors af-
fecting them. As more components are investigated, the
system is conceptually reassembled. Process studies have
employed many research approaches. To investigate causes
for landsliding, for example, descriptive case studies have
detailed the characteristics of specific slides (Fredriksen
1963); extensive surveys have identified variables associ-
ated with landslide distribution (Furbish and Rice 1983,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Rice and Lewis 1986); laboratory experiments have revealed
the effect of soil moisture on shear strength (Yee and Harr
1977); mathematical models have explored the mechanisms
that trigger slides (Swanston 1970); and monitoring pro-
grams have tracked pore pressures at likely slide sites
(McGreal and Craig 1977).
Some system responses are governed by interactions
between several components, and studies of isolated elements
may not provide enough information to understand the
response. In these cases, process studies are expanded to
focus on interactions instead of on events or species. Predicting
a channel’s morphological response to flooding, for example,
requires understanding its response both to altered sediment
load and to altered flow, and also to the combined effects of
sediment and flow. Physical and biological components also
affect each other, so both geomorphological and ecological
process research require knowledge of the other discipline.
Watershed Case Studies
Most research that documents CWEs has been in the form of
case studies. An impact is observed, the situation is de-
scribed, and the causes are investigated. This procedure de-
fines the issues and types of mechanisms involved, and has
traditionally been the first step in exploring a new type of
problem. Quantitative results of case studies are not general-
izable, but the processes and effects they describe often are.
Watershed case studies usually include reconstruction of
original watershed conditions and documentation of changes.
Changes are evaluated either by monitoring ongoing responses
or by finding evidence of intermediate conditions, and aerial
photographs often provide an important information source.
Effects of particular land-use activities are estimated or
measured, and these provide a link between land use and
watershed response. Without this information, association
between land use and watershed response is only
circumstantial.
G.K. Gilbert (1917) provided the first comprehensive
analysis of CWEs in a case study of hydraulic mining impacts
in the Sacramento River watershed. Gilbert’s circumstantial
linkages between land use and impacts were strengthened by
field measurements and by his understanding of component
processes. Subsequent CWE studies followed a similar pattern
by combining historical reconstructions with process analyses.
Collins and Dunne (1989), for example, used this approach
to evaluate cumulative changes from gravel mining in rivers
of the Olympic Peninsula, Washington.
Watershed experiments may also be case studies. Basin
attributes are monitored to describe the system, changes are
made, basin response is monitored, and pre- and post-change
records are compared to define the magnitude of the response.
This method ensures that attributes such as bedrock and
topography are identical for both states, but it cannot account
for differences in temporal variables. Occurrence of a major
storm during one study period, for example, may either alter
the watershed’s response to manipulation or introduce changes
17
that cannot be distinguished from treatment effects. Similarly,
changes caused by a fortuitous natural landslide may
overshadow the effects of land use. Case studies can describe
a particular basin’s response to conditions it has experienced,
but results cannot be generalized unless the response mechanism
is understood.
Paired- and Multi-Basin Studies
Many watershed experiments manipulate one basin and pre-
serve a second as a control. This strategy accounts for tem-
poral variations, but differences in basin characteristics may
affect the basins’ responses, and occurrence of a landslide or
other natural event in one basin confuses inter-basin com-
parisons. The effects of these complications can be esti-
mated only if response mechanisms are understood. Some
paired-basin studies are descriptive rather than experimental
and compare conditions in existing pre- and post-distur-
bance watersheds.
Paired-basin studies produce site-specific results that
cannot be generalized without other information. It is difficult
to recognize and control for all potentially important variables
in these experiments, and linkages between imposed change
and observed response are often treated as a black box,
further hindering interpretation. Nevertheless, these studies
define the magnitudes of potential responses and often identify
the most significant problems for further study.
Multivariate statistics often can sort out the relative
importance of variables if multiple basins are compared, and
these results are more readily generalized. However, any
collection of sample watersheds includes many variables. If
these are to be evaluated adequately, then the sample set
must include the likely range of each variable, the variables
must be independent, and enough samples must be evaluated
to provide statistically sound results. Multi-basin comparisons
are also useful in settings where no undisturbed watershed is
available for paired-basin comparisons. In this case, multiple
basins are selected that represent a gradient of treatment
intensities. If the basins’ responses show a trend that correlates
with the treatment intensity, then an association between
treatment and response can be inferred even in the absence
of a standard “control” watershed.
Multiple-basin comparisons can identify associations
between watershed treatment and response and can reveal
patterns of occurrence, but they cannot demonstrate causes.
These studies may lead to hypotheses of why a particular
response occurs, but the evidence they provide in support of
those hypotheses is circumstantial. Callaham (1990) discussed
practical aspects of carrying out paired and multi-basin studies.
The System as Focus
Researchers have recently begun to study watersheds as
coherent systems. Tangible results from these studies have
so far been few, because appropriate methods are still being
18
developed. Some such methods have arisen from general
systems theory (Von Bertalanffy 1968), a field that studies
the properties of systems. According to systems theory, a
system’s behavior is determined by the interactions com-
prising it, so its response might be predicted from knowl-
edge of its structure. Geomorphologists have translated geo-
morphological ideas into the vocabulary of systems theory
(e.g., Chorley and Kennedy 1971), and this provides a useful
way to describe interactions, but it has not yet produced
major advances in understanding watersheds.
Recent developments in chaos theory (Prigogine and
Stengers 1984) aid a watershed-level approach by classifying
and explaining styles of response that might be generated by
complex systems. Chaos theory has not yet been applied to
watershed studies, although geomorphologists occasionally
use its concepts to interpret study results.
The discipline of landscape ecology (Naveh and Lieberman
1984) is most widely recognized in Europe. Landscape ecology
views landscapes as collections of interdependent physical
and biological processes, so studies of landscape-scale
problems incorporate both geomorphological and ecological
components. This approach has occasionally been adopted
in the past (e.g., Hack and Goodlett 1960) and does not
represent a new field, but its formal recognition will encourage
interdisciplinary work and legitimize landscape-scale studies.
Problems are also being explored at the scale of watersheds.
Parsons (1982) defined basin-wide trends in hillslope
morphology, for example, and Benda and Dunne (1987)
found that debris-flow mobility could be predicted from
channel network geometry. Methods are being developed to
characterize watersheds using information gathered at a
process scale. For example, sediment budgeting provides a
framework quantifying rates of sediment production, transport,
and redeposition throughout a basin (Dietrich and others
1982). These values can be combined according to the
distribution of rate-controlling variables to estimate process
rates and distributions in the sampled area. Processes are
influenced by changes in land use and environmental
parameters, and sediment budgeting allows estimation of the
effects of such changes on basin response (Reid 1989, 1990).
Work is also being done to incorporate spatially explicit data
into sediment budgeting (e.g., Reid 1989), and this will
eventually enable sediment budgets to be used to model
landform evolution and spatially distributed responses to
environmental change.
Recent advances in computer software and remote sensing
also aid watershed-scale studies. Geographic Information
Systems (GIS) now provide an efficient way to organize,
manipulate, and evaluate spatial data. GIS is being used in
watershed simulation models and sediment budgeting to
produce spatially distributed results. New methods of remote
sensing and image analysis permit rapid mapping of watershed
attributes over large areas in a form that can be incorporated
into GIS databases. This combination allows rapid
characterization of watersheds for modeling.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Chapter 4
Assessment of Cumulative Watershed Effects
T he ultimate goal of CWE research is to be able to
predict the effects of multiple land-use activities. If the
magni-tude of a physical or biological change can be predicted,
then its potential economic, philosophical, and political
significance can be more easily assessed. CWEs have been
described and evaluated for many years, but under different
labels. This chapter examines published descriptions of CWEs,
CWE assessments, and methods of predicting CWEs.
Literature specific to CWEs is not extensive, but many
other studies examine the effects of land-use activities on
environmental characteristics, the response of watershed
function to environmental changes, and the impacts of altered
watershed function on resources and values. This work is
described in following chapters, and is the foundation for
most of the CWE research outlined in this chapter.
Types of Cumulative Watershed
Effects
Most watershed impacts can be evaluated as cumulative
effects because they occur in the context of multiple land-
use activities. CWEs can be generated either at the site of
land-use disturbance or downstream.
On-Site and Off-Site CWEs
Environmental changes can cause on-site CWEs if they
persist long enough for sequential or progressive effects to
accumulate, or if changes generated elsewhere are transported
to a site and interact with on-site changes. Recognized on-
site CWEs include impacts on soils, nutrient cycling, and
soil moisture recharge. These changes affect long-term
productivity, and can have major impacts on timber
management, range use, and agriculture.
Off-site CWEs assume many forms, but each occurs
because environmental change alters the production or
transport of watershed products (water, sediment, organic
material, chemicals, and heat). Off-site CWEs can occur
when processes are altered for long enough that changes can
accumulate through time, when responses from multiple
sites are transported to the same site, or when a transported
response interacts with an on-site change at another site.
Altered hydrology can modify the time distribution and
amount of runoff and thus change baseflows, peakflows, and
flow seasonality. These changes can then alter flood
frequencies; aggrade, incise, or widen channels; and change
the size distribution of stream-bed sediment. Changes in
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
sediment input can trigger similar responses through different
mechanisms. Altered flood frequencies and channel
morphology affect flood-plain land use, fisheries resources,
and navigation.
Changes in the input and transport of organic material are
important both because large organic debris affects channel
morphology, and because organic matter is a food source for
aquatic animals. Off-site impacts from altered basin chemistry
generally appear as water quality problems affecting aquatic
biota and water supplies. Changes in stream temperature are
important primarily because of their effects on biota.
Interactions That Generate CWEs
A single type of land-use activity can generate CWEs if it
occurs repeatedly or persistently at a site or if it occurs over a
progressively larger area. Different land-use activities can
produce CWEs if they occur sequentially or if they coexist
within a watershed. Impacts are cumulative when activities
reinforce the same watershed response, when multiple re-
sponses disturb the same resource, when one response pro-
vokes another, or when responses interact to produce another.
Repeated, progressive, sequential, or coexisting activities
can cause CWEs if the activities cause the same type of
environmental change. For example, logging, road use, and
grazing all compact soil, and the total area compacted in a
basin may include contributions from each activity. Because
compaction can be persistent, sequential performance of
these activities at a site can cause on-site cumulative
compaction. In these cases, both activities (A and B in the
following examples) cause the same type of change to an
environmental parameter (Y), and the resulting CWEs (Z)
can be evaluated by understanding the extent of the change
in Y. Activities A and B contribute to the same type of
impact whether acting alone or together (fig. 3a).
CWEs can also occur when different activities affect the
same resource by different mechanisms but contribute to the
same response; these are called complementary effects.
Increased water pollution and increased fishing pressure
both decrease salmon populations, so the overall population
decrease is a cumulative result of both mechanisms. Similarly,
flooding might be aggravated both by increased quickflow
runoff and by channel diversion. Complementarity implies
that activity A causes Y1, B causes Y2, and both Y1 and Y2
contribute to cumulative impact Z. Either activity acting
alone also contributes to Z (fig. 3b).
Both mechanisms described above involve changes that
can be recognized by examining the effects of each land use
independently. CWEs can also result from responses that interact
in ways not predictable from analyses of individual effects.
CWEs can be generated by cascading influences, where one
19
type of use influences a second to provoke a CWE. For example, TYPE OF CUMULATIVE
EFFECT
EFFECTS OF EFFECT OF COMBINED
INFLUENCES INFLUENCES
EXAMPLE

lack of shade after clearcutting might concentrate cattle in ACTING ALONE

road
riparian buffer zones, thus increasing bank erosion rates; or A. Same-influence A B A B logging traffic
effects
construction of logging roads might improve access to an area
Y Y compaction
and increase recreational impacts. In these cases, activity A Y

modifies activity B to cause impact Z, and Z does not necessarily Z Z Z altered peak flow
occur unless both A and B are present (fig. 3c).
In other cases, impacts result from interactions between B. Complemntary A B A B logging
road
effects building
different environmental changes, and influences are
diverted
interdependent in causing the impact. For example, toxic Y1 Y2 Y1 Y2 compaction
channels
compounds may form when two introduced chemicals mix.
Z Z Z
CWEs can also occur if one type of change alters the altered peak flow

importance of another. For example, hatchery-related

C. Cascading A B road recreational
introduction of Bacterial Kidney Disease to wild fish effects building use
populations may decrease the ability of wild stocks to survive
Y1 Y2 increased increased chemical
environmental change. In these cases, response Z happens access & nutrient inputs

when both activities A and B occur. Either A or B acting Z impaired

alone does not necessarily provoke Z (fig. 3d). water quality

D. Interdependent A B A B industrialization mining

effects

Y1 Y2 Y1 Y2 input of input of
chemical 1 chemical 2

Retrospective Assessments of Z formation of chemical 3

Cumulative Watershed Effects
Most published CWE studies evaluate the reasons for ob-
served problems. The impact has already occurred and is
well defined, so the scope of the problem and the possible
controlling elements are known. This problem is less com-
plex than that of predicting possible CWEs from unspecified
land-use combinations. Retrospective studies evaluate sce-
narios that are already complete, and this is more attractive
to researchers than prognoses of future change. Explana-
tions of an existing condition are immediately testable; pre-
dictions of the distant future are not.
Examples of Retrospective Analyses
Four types of land-use combinations (repeated, progressive,
sequential, and coexisting activities) can generate CWEs by
four types of mechanisms (combinations of the same, comple-
mentary, cascading, or interdependent influences). These
combinations and mechanisms form a 4x4 matrix that is a
convenient framework for describing CWE studies (table 2).
Same-Influence Effects
If an environmental parameter is altered in the same way by
repeated or multiple activities, those activities all contribute
to the watershed’s response, and response is a cumulative
effect of the activities. The change may impact a resource
directly or alter watershed processes and cause an indirect
impact, but evaluation is relatively straightforward in either
case. If the general relationship between an environmental
change and an impact is known, then the effect of a land-use
activity can be estimated by evaluating its effect on environ-
Figure 3—Combinations of influences that can generate CWEs.
Types are described in the text. A and B are activities, Y is an
environmental parameter, and Z is an impact.
mental parameters. These effects often are easily measured.
Repeated or prolonged activities can augment an impact
by repeatedly perturbing the same environmental parameters.
Collins and Dunne (1989) examined air photos, gauging
records, and mining records to demonstrate that streambeds
downstream of mined gravel bars are incising at rates that
account for the cumulative gravel deficit left by mining.
Sediment budgets suggest that incision will continue if gravel
mining continues at present rates. Froehlich (1979) and Geist
and others (1989) found that soils required long periods after
logging to recover from compaction, and inferred that recovery
may not be complete before the next logging cycle. Compaction
may thus accumulate through multiple cutting cycles.
In the case of gravel mining, the activity occurred annually,
so demonstration of a CWE required evidence of continued
change. In contrast, logging was to be repeated at long intervals,
so the recovery period for compaction had to be shown to be
longer than the logging rotation. Each study used surrogate
evidence to measure response trends through time. Collins
and Dunne (1989) used historical air photos, while Froehlich
(1979) and Geist and others (1989) compared soil densities
on sites logged at different times. Monitoring of changes
through time would have provided the same information, but
studies would have required a half-century of monitoring.
Both of these problems were approached at a process scale,
and researchers focused on mechanisms driving the changes.
Activities that progress through an area progressively
broaden their effects on environmental parameters. CWEs
from a progressive activity can be demonstrated by comparing

20 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Table 2—Examples of studies diagnosing cumulative effects. Brackets indicate hypothetical or undocumented examples.

Activity type Same influence Complementary influences Cascading influence Interdependent influences

Activity repeated Repeated gravel mining lowers the channel
or progresses bed (Collins and Dunne 1989)
through time
Repeated logging progressively compacts
soil (Geist and others 1989, Froehlich 1979)
Activity Amount of fine sediments in redds increases
progressive with road density (Cederholm and Reid
through area 1987)
Peak flows increase as logged area increases
(Dietterick and Lynch 1989)
Channel form changes because woody [Traffic degrades ORV track and alters runoff Acid and arsenic from mining combine
debris is destabilized by logging and future over widening area] (Mok and Wai 1989)
inputs will be low (Hogan 1987)
Decreased flows and sediment alter bed
(Pemberton 1976)
Streams incise when overflow is restricted [Erosion severity increases disproportionately
and flow increases (Nanson and Young as logging progressively enters steeper areas]
1981)
Increased sediment and flows from logging
alter streams (Lyons and Beschta 1983)
Channel change increases with mining
(Touysinhthiphonexay and Gardner 1984)

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

Urban runoff and conveyance enlarge
channels (Hammer 1972)
Channel change increases with logging
(Ryan and Grant 1991)
Sequential Logging, burning and herbicide increase Fire risk increases from grazing after
activities nitrogen loss (Feller 1989) logging (Zimmerman and Neuenschwander
1984)
Logging and leaching deplete magnesium
supply (Roberts and others 1989)
Coexisting Cropping and grazing increase Fishing and habitat loss reduce fish
activities sedimentation (Reid 1990) populations (Salo and Cederholm 1981)
Reduced flow and high sediment aggrade
a stream (Petts 1984)
Aggradation and channelization increase
floods (Gilbert 1917)
Wetland conversion lowers water quality
(Johnston and others 1988)
[New Melones Dam destroys river run,
increases crowding nearby]
[Construction of logging roads improves access
and leads to increased recreational impact]
[Urbanization is accompanied by increased
recreational demands, increasing recreational
impact]
U.S. demand for beef encourages conversion of
rainforest to range (Miller and others 1991)
Socio-economic conditions lead to grazing
practices that accelerate Tanzanian soil
erosion (Reid 1990)
Grazing and burning alter vegetation
(Collins 1987)
Vegetation change makes forest
susceptible to smog (Mazurski 1986)
Conservation after erosion forms gullies
(Costa 1975)
Disease introduced from hatcheries
decreases ability of fish to survive
environmental stress (J. Nielsen, personal
communication)

21
impact magnitudes at different stages in its progress. Dietterick
and Lynch (1989) measured changes in peak discharges as
clearcutting progressed through watersheds. Incremental
increases were greater than those predicted from the area
cut, implying a synergistic interaction.
Cederholm and Reid (1987) described a series of studies
that associated spawning gravel siltation with percentage of
a basin in roads. The relationship was developed from
measurements of fine sediments in gravels in many watersheds
with different road densities. However, statistical association
does not demonstrate causality, and a sediment budget was
used to identify logging roads as a dominant sediment source.
Results were translated to an impact on fisheries using
laboratory and field measurements of survival-to-emergence
as a function of siltation level.
CWEs from sequential activities are demonstrated by
measuring impact levels after each activity. Feller (1989)
measured nitrogen loss by streamflow in catchments that
had been clearcut, clearcut and burned, and clearcut with a
follow-up herbicide application. He found that herbicide use
mobilizes more nitrogen than slashburning, and that addition
of either of these treatments produces more than logging
alone. Measurements were continued over a 5-year period to
assess recovery rates. Causes could not be demonstrated
because only outflow was measured, but they could be inferred
because earlier process-based studies examined nutrient loss
from each of these activities. Without this background of
research, results from a study of this type would be difficult
to interpret or generalize.
CWEs are relatively easy to evaluate at a process scale
when coexisting activities cause the same type of
environmental change: each environmental change can be
assessed independently and the changes summed to estimate
their combined effect. Reid (1990) used this approach to
estimate sediment yield from multiple land-use activities in
Tanzania. Soil-loss rates on each type of surface were
estimated from published relations and field measurements,
and loss rates were summed according to the distribution of
land use to estimate total input from the combination of
land-use activities.
Complementary Influences
Complementary effects occur when land-use activities con-
tribute to the same result through different mechanisms, and
their evaluation requires that the link between altered envi-
ronmental parameters and watershed response be established
for each activity. As a result, few evaluations of complemen-
tary effects are “complete”: rarely does a study demonstrate
every link in an association, and most use results of earlier
studies to infer some linkages.
Hogan (1987) evaluated the effect of logging on the
frequency of debris jams on the Queen Charlotte Islands,
British Columbia. His measurements and estimates of debris
input, storage, and output showed that debris is less stable
after logging than before. Inputs of logging slash and decreased
channel stability both contributed to this change, and future
22
logging cycles will prevent regrowth of large trees and ensure
that organic debris inputs continue to be mobile. This study
measured the immediate complementary effects of logging
and inferred future effects as logging is repeated in the
future. In contrast, Pemberton (1975) provided a complete
assessment of one type of complementary effect from the
operation of Glen Canyon Dam. The dam decreased both
downstream sediment load and peak discharge, and both
effects contributed to channel-bed coarsening. Pemberton
used bedload equations that take both influences into account
to successfully predict the observed changes.
Progressive activities that cause complementary effects
are widely studied, but the component influences are usually
not disentangled. Touysinhthiphonexay and Gardner (1984),
for example, surveyed channels in 29 watersheds that had
different proportions of their area strip-mined and defined a
relationship between channel destabilization and proportion
mined. Channel enlargement was attributed to the combined
effects of increased peak flows and increased sediment input,
because earlier studies had shown that both of these changes
can alter channel morphology. Hammer (1972) used a similar
approach to explore the effects of progressive urbanization
on channel form. Hammer attributed changes to increased
peak flows from altered drainage networks and increased
runoff but did not evaluate the relative importance of these
causes. Ryan and Grant (1991) measured the disruption of
riparian vegetation on sequential aerial photographs in basins
that had been logged to different extents and showed a
correlation between riparian area disturbed and basin area
logged. Lyons and Beschta (1983) associated channel changes
in the Middle Fork Willamette River with increased sediment
loads and peak flows caused by timber management and
roads. Progressive changes were measured using sequential
aerial photographs and stream gauge records, and the effects
of altered sediment and flow were inferred from earlier
process studies. Nanson and Young (1981) compared
urbanized and undisturbed channels in southeastern Australia
to demonstrate enlargement associated with urbanization.
They attributed changes to increased peak flows from the
growing impervious area and from levee construction that
prevented storage of water on floodplains. The progressive
nature of the change was inferred from the effects of increasing
urbanization on the component mechanisms.
Sequential uses can also produce complementary effects.
Zimmerman and Neuenschwander (1984) compared vegetation
in grazed and exclosed selection cuts to measure the effects
of grazing on fuel loading after logging. Cattle removed
cool-burning, herbaceous fuels while increasing the volume
of downed wood, a hot-burning fuel, and thus increased fire
hazard and decreased the effectiveness of controlled burns.
These changes compounded the effects of logging, which
had increased fuel loading and provided routes for ground
fires to reach the forest canopy by promoting the growth of
spindly, sub-canopy trees. Roberts and others (1989) associated
the decline of some European forests with magnesium
deficiencies produced by the sequential effects of (1) export
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
of nutrients in logs, (2) leaching from acid precipitation, and
(3) increased magnesium demand as second-growth forests
reached a growth stage requiring large nutrient inputs.
Salo and Cederholm (1981) presented a clear analysis of
complementary influences in an evaluation of the effects of
coexisting land-use activities on coho populations in the
Clearwater River, Washington. The authors used experimental
results, field measurements, and fishery records to estimate
the effects of road-related redd siltation, logging-related
habitat modification, and fishing pressure. Each of these
land uses increases fish mortality through a different
mechanism, so incremental impacts could be summed to
estimate their combined effect. Gilbert (1917) described
complementary effects from coexisting land uses in the
Sacramento River basin. Construction of flood-control levees
increased flood peaks by reducing floodplain storage, while
upland hydraulic mining increased sediment input and
aggraded the channels. Both effects increased downstream
flood risk. Gilbert inferred causes for observed changes
using his understanding of river mechanics and results of a
sediment budget he constructed for the watershed. Petts
(1984) described channel aggradation from the combined
effects of reduced sediment transport capacity below a dam
and increased sediment input from silvicultural site preparation
on a tributary. Aggradation occurred only at the mouth of the
affected tributary, where the altered flow regime could not
remove the additional sediment. Johnston and others (1988)
used a geographic information system to evaluate the
combined effects of wetland loss and increased pollution on
water quality near Minneapolis.
Because of the complexity of the interactions involved,
most studies of complementary influences draw inferences
from prior studies or basic principles, and thus produce
qualitative results more readily than quantitative ones. Effects
may be additive if different mechanisms do not interact, but
many changes do affect each other. For example, increases
in both sediment production and flood peaks can destabilize
channels, and increased peak flows also usually increase
sediment input, so the combined effect cannot be predicted
simply by adding the expected response to each influence. In
addition, responses may not be linear. Touysinhthiphonexay
and Gardner (1984) found that combined increases in peak
flows and sediment input gradually enlarge channels up to a
certain impact level, but then a major shift in channel stability
occurs. The existence of a threshold cannot be predicted by
adding incremental changes from peak discharges to those
from sediment loading unless the mechanisms producing the
threshold response are understood.
Cascading Influences
CWEs also occur when one activity influences another to
provoke a watershed response. Miller and others (1991) de-
scribed such a case in the Brazilian rainforest, where defores-
tation contributes to loss of soil productivity. In this case,
deforestation is in part caused by conversion of forest to
rangeland, and an underlying cause for the conversion is the
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
demand for cheap beef in North America. Analysis of the
magnitude of future impacts from deforestation would thus
require forecasts of beef demand in international markets.
Land-use impacts are usually evaluated in terms of their
immediate cause, so the nature of the interaction between the
land-use activities is rarely examined even in cases where an
impact results from a cascading influence. Studies of cascad-
ing influences are thus uncommon, but hypothetical examples
are described in table 2 to illustrate potential influences.
Repeated activities or those that continue through time
may produce a cascading influence if early use influences
how later use is carried out. For example, repeated use of a
muddy trail or off-road vehicle track eventually produces a
mire and forces later traffic to broaden the impacted area by
forming parallel tracks. An activity that progresses through
an area can also modify its impact as it proceeds. Many types
of land use begin on optimal sites and spread to more sensitive
ones. Disproportionate responses from sensitive sites then
cause the overall impact to increase more rapidly than the
proportion of area affected.
An activity can also influence subsequent use if it alters
the environmental parameters that control the second use.
For example, construction of New Melones Dam destroyed
a popular river run as demand for whitewater recreation was
increasing in central California. Impact intensity on the
remaining whitewater runs in the area increased due to the
decreased resource availability. Road construction also
frequently produces cascading impacts: roads constructed
for timber management open areas to recreational access
and lead to increased recreational impact.
Coexisting activities can influence one another in a similar
way. For example, recreational demand increases as
urbanization proceeds, increasing all types of recreational
impacts in nearby wildlands. Similarly, Reid (1990) attributed
high soil erosion rates on Tanzanian rangelands ultimately
to the socio-economic conditions in the area that inadvertently
promote high stocking rates.
In each of these cases, observed impacts could be evaluated
simply as results of the activity that directly caused them:
campsite degradation could be analyzed as a function of
recreational pressure without acknowledging the role of dam
construction or increased urbanization, and timber-
management impacts are usually expressed as a per-unit-area
average rather than accounting for changes in site sensitivity
as logging progresses. Prediction of future impacts, however,
would require a full understanding of the cascading influences.
Interdependent Influences
Influences are interdependent when changes interact to create
an impact or when one change alters the importance of another.
An interaction between introduced chemicals is a clear example
of an interdependent influence. Mok and Wai (1989), for ex-
ample, described chemical changes caused by an Idaho mine:
mine dumps leached arsenic into streams while mine drainage
increased stream acidity, and acidic conditions decreased the
impact of arsenic by altering it to a less mobile form.
23
 Mazurski (1986) inferred interdependent sequential
influences to explain forest dieback in the Sudetes Mountains
of Poland. Five centuries of logging depleted hardwoods in
montane forests and replaced them with conifers, which are
more sensitive to acidic conditions. Industrial smog then
acidified the environment enough to injure conifers. Costa
(1975) found that high sediment loads in streams of the
southeastern United States reflect the sequence of land use
there: sediment eroded during two centuries of intensive
agriculture accumulated in valleys, but channels began to
incise and export the sediment when declining agricultural
use decreased incoming sediment loads. In both of these
cases, neither of the most recent changes operating alone
would have produced the present impacts. Collins (1987)
documented another example of interdependent sequential
influences by evaluating community composition on natural,
grazed, burned, and grazed and burned prairie plots. Grazing
was found to increase species diversity on burned plots, while
burning reduced species diversity on ungrazed plots. Interaction
between disturbances produced the highest diversity.
Coexisting land-use activities also generate interdependent
effects. For example, Bacterial Kidney Disease has been introduced
to wild anadromous fish populations in California from hatchery
stock. Although the disease does not kill the fish directly, it makes
them more sensitive to environmental stress, and so less able to
survive the types of environmental change induced by other land-
use activities (J. Nielsen, USDA Forest Service, Pacific Southwest
Research Station, personal communication).
Interdependent influences must be evaluated together to
assess the nature of their interdependency. The processes of
interaction often control the nature and occurrence of
watershed or population response.
Patterns of Retrospective
Understanding
Most research approaches described in Chapter 3 have been
applied to CWE studies (table 3). In addition, some studies
used published equations to calculate process rates, others
used existing evidence for past conditions (e.g., air photos)
or surveyed sites of a variety of ages, and several predicted
the nature of effects using a basic understanding of how the
component processes work. No method dominated among
the studies surveyed, and most used multiple approaches.
Most examples either used process studies to define impacts
or used the results of earlier process studies to infer the causes
of impacts (listed as R in table 3), and the importance of
process studies increased with the complexity of interactions
causing the impacts. Key research for interpretation and
assessment of CWEs thus is not limited to studies that focus

Table 3—Approaches used by studies described in table 2 to diagnose CWEs. Studies are listed in the same order as in
table 2, and symbols are explained below.

Same Complementary Cascading Interdependent

influence influences influences influences
ADEMQRST/BCHMP ADEMQRST/BCHMP ADEMQRST/BCHMP ADEMQRST/BCHMP

Activity A.EM..S./.C..P .D...R../B...P .....R../..H.. .DE...S./.C..P

repeated A....RS./...MP ...M..../.C..P
A....R../...MP

Activity ..E..RS./...MP .D....S./...M. .....R../..H..

progressive ..E.Q..T/B.... AD....../.C...
....Q.S./...M.
A...Q.S./...M.
AD...RS./...MP

Sequential ..E....T/B.... AD...R../B...P .....R../..H.. ..E...../...MP

activities .....R../..H.P .....R../..H.P
AD....S./.C..P

Coexisting A..M..S./.C..P ..E..RS./..H.P .....R../..H.. .....R../..H..

activities .D...R../.C... .....R../..H..
AD.M.R../.C..P .D...R../.C...
....Q.S./...M.

Approach or method Scale

A reconstructs past using air photos or B paired basins
records, or by sampling age distribution C case study
D based on descriptive measurements H hypothetical or general case
E experimental M multiple basins
M modeling or calculated results P process study
Q based on statistical analysis
R reasoned from basic principles
S survey of multiple sites
T long-term temporal monitoring study

24 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

specifically on CWEs, but also includes basic research on
geomorphological, hydrological, and ecological processes; many
of these background studies are described in Chapters 5 and 6.
Most studies that compared watershed-scale responses
sampled more than two basins so that the cumulative nature
of a response could be demonstrated. Both the studies that
employed temporal monitoring compared paired sites, but
paired basins were otherwise rarely used.
Influences are more easily identified than quantified. Many
of the cited examples produced only qualitative results because
they depended on inference or on general process information,
and many were intended merely to indicate how an effect
occurred. Even when results were quantified, the relative
contributions of particular influences often were not.
It is easiest to quantify impacts from coexisting and
progressive land uses that have the same effect on
environmental parameters (table 2, column 1, rows 2 and 4),
because individual effects can often be summed and their
combined effect evaluated. This approach may also be applied
to complementary effects (table 2, column 2) if influences do
not interact with one another. In contrast, responses to repeated
and sequential use often change as an impact progresses. For
example, soil compaction increases with increased traffic, but
most compaction occurs with the first few passes. Compaction
from 100 passes thus cannot be calculated as 100 times the
compaction from a single pass. The effects of cascading and
interdependent influences can be evaluated and interpreted
only if the interactions causing them are understood, so
assessment requires knowledge of the processes involved.
Prediction of Cumulative
Watershed Effects
Research studies rarely focus on predicting future effects be-
cause predictions are not testable in the framework of the
study, and predictions described in published research usually
are by-products of present effects. However, many predictive
CWE assessments are carried out by nonresearchers in sup-
port of particular land-use plans. Federal and State agencies
are required to predict CWEs for planned projects, as are
many private developers. These reports are often in the form
of environmental impact statements, environmental assess-
ments, and timber harvest plans, and they do not constitute
research. Other predictive studies include proprietary consult-
ing reports that detail anticipated impacts from planned projects.
Examples of Predictive Studies
The few predictive studies available can be organized in the
same way as retrospective studies (table 4). None of the de-
scribed studies evaluates cascading or interdependent effects.
Same-Influence Effects
Interactions between processes rarely need to be evaluated
when impacts are caused by alteration of a single type of
parameter. As described earlier, however, the response

Table 4—Examples of predictive cumulative effect studies

Activity Type Same Influence Complementary Influences

Activity Effects of soil conservation on productivity Effects of dam construction on

repeated or (Benson and others 1989) delta erosion (Pickup 1980)
progressive
through time Effects of diverted flows on channel
Effects of erosion on productivity form (Jackson and Van Haveren
(Christensen and McElyea 1988) 1987)
Effects of repeated logging on productivity Effects of cloud seeding on channel
(Waide and Swank 1976) form (Rango 1970)

Effect of increased flow on fish

production (Johnson and Adams 1988)

Activity Effects of tidal marsh reclamation Effects of urbanization on local

progresses on aggradation of bay-mouth bar climate (Lein 1989)
through area (Gilbert 1917)
Effects of ozone depletion on
agricultural production (Adams and Rowe 1990)
Effects of logging pattern on
channel stability (Ziemer and others 1991)
Sequential [Predictions of the effect of
activities planned mitigation works]
Coexisting Effects of land uses on estuary Effects of riparian vegetation and
activities sedimentation (Dickert and Tuttle 1985) land use on fish production (Theurer and others
1985)
Effects of pesticides and sediment
on fish production (Braden and
others 1989)

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 25

generated by an incremental environmental change often
changes as an impact develops.
Many recent studies have evaluated the cumulative impact
of progressive erosion on crop yields. Benson and others
(1989) used a published model (EPIC: Erosion Productivity
Impact Calculator) to predict the long-term effects of erosion
on soil productivity, while Christensen and McElyea (1988)
developed a procedure for calculating marginal changes in
productivity using field measurements. In these cases,
cropping is repeated through time, and cumulative effects
are on-site. Waide and Swank (1976) modeled the cumulative
impact of multiple logging cycles on nitrogen budgets and
forest productivity in a southeastern forest.
Gilbert (1917) predicted the effects of progressive marsh
reclamation on shipping through the Golden Gate by
calculating the resulting change in volume of the tidal prism,
and thus of the transport capacity across the bay-mouth bar.
These calculations were based on fundamental research on
sediment transport mechanics.
No published examples predicting the effects of sequential
uses were found, but these are commonly carried out during
planning of rehabilitation and mitigation projects. The effects
of current practices are usually estimated with reference to
undisturbed conditions, and changes expected from the
planned work are then predicted. Dickert and Tuttle (1985)
calculated percentage of bare ground for coexisting land
uses in the Elkhorn Slough watershed and used these as
erosion indices to aid in planning mitigation work to reduce
estuary sedimentation.
All studies in this category employed process-based
analyses. Prediction implies that conditions or results change
through time, and processes are often altered as they progress.
Where processes are not altered, results are calculated by
summing the process rates produced by each activity (e.g.,
Gilbert 1917). In contrast, where progress of a change modifies
its rate, the marginal increase in impact due to an incremental
change in driving parameter must be evaluated either
empirically (e.g., Christensen and McElyea 1988) or through
a basic understanding of the processes involved (e.g., Benson
and others 1989).
Complementary Influences
Most predictive studies concern land-use activities that alter
multiple environmental parameters and watershed processes,
and changes may be modified by interactions with one an-
other. These changes can cause complementary impacts if
they affect a target resource through different mechanisms.
Interactions must be evaluated if complementary impacts are
to be predicted, but interactions are often complex, and pre-
dictions of complementary effects are usually qualitative.
Pickup (1980) used a flow-routing model to predict the
flow regime resulting from a proposed dam in New Guinea.
He then used a basic understanding of river mechanics to
assess qualitatively the river’s long-term response to the new
regime and to the accompanying decrease in sediment input.
Jackson and Van Haveren (1987) addressed a similar
26
problem: they needed to predict the effects of a proposed
Alaskan flow diversion on channel morphology. The
researchers used a regional relation between flow
characteristics and channel geometry to predict the channel
form that would be in equilibrium with the proposed flows.
Rango (1970) used a similar approach to estimate the effects
of cloud seeding on channel geometry in the western United
States. In both cases, flow was to be altered without a major
change in sediment input, but the altered flows would modify
bank erosion rates and indirectly change sediment input. The
studies implicitly accounted for these adjustments by using
statistical descriptions of equilibrium channel forms.
Johnson and Adams (1988) also evaluated the effects of
changing flow regime, but they focused on the impact to a
particular resource, a recreational steelhead fishery. They
used a published stock-recruitment model to estimate the
number of returning adults for a given smolt production rate,
and used measurements of juvenile survival as a function of
flow level to assess smolt production. Impact magnitudes
could then be estimated by calculating the number of returning
adults as a function of flow regime. A questionnaire that
measured anglers’ valuation of the resource was used to
translate impact magnitude into a dollar value. This approach
allowed both the biological magnitude of changes and their
economic significance to be estimated, but did not address the
existence value of steelhead to other sectors of the population.
In each of these cases, cumulative effects were expected
from continuing activities that alter channel flows. A one-
time change in flow is similar to an isolated storm or drought
and usually has only a short-term effect, but a chronic alteration
cumulatively alters channel morphology and biological
communities to reflect a long-term change in hydraulic regime.
The cumulative impacts of activities that progress through
an area are the focus of many predictive studies. In this case,
too, influences are usually complementary and interacting.
Lein (1989) used an energy balance model to predict local
increases in average temperature of 0.16 °C from urbanization
in Ohio. The study accounted for the effects of altered
albedo, surface roughness, and evapotranspiration rates on
six types of land cover. Records of changes for 1969 to 1979
were used to calibrate the model and to identify land-use
changes that accompany urbanization.
Adams and Rowe (1990) predicted the economic impact
of stratospheric ozone depletion on agricultural production
in the United States. Ozone depletion reduces crop production
by increasing ultraviolet-B (UV-B) radiation and tropospheric
ozone, and altered yields were calculated from existing
information on the relations between pollutant inputs and
ozone depletion, ozone depletion and UV-B penetration,
ozone depletion and tropospheric chemistry, UV-B and crop
yield, and tropospheric ozone and crop yield. Finally, an
economic model used yield estimates to predict annual losses
of $1.3 to $2.5 billion for a 15 percent decrease in stratospheric
ozone. The researchers emphasized that results are preliminary
estimates, and that accurate prediction requires research to
better define the relationships employed by the model.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Ziemer and others (1991) combined equations describing
landslide probabilities, sediment transport, scour depth, and
channel morphology to construct a model that predicts the
effects of logging patterns on channel stability. The model
was presented primarily as a heuristic tool for indicating
qualitative trends.
Theurer and others (1985) predicted the effect of altered
coexisting land uses on fish. The Tucannon River basin in
Washington includes grazed and cropped land, and riparian
revegetation is planned for salmon-producing reaches to
mitigate temperature increases. The study used an existing
temperature model in conjunction with models of water and
sediment yield (Garbrecht and DeCoursey 1986), sediment
intrusion into gravels, dissolved oxygen, and juvenile survival
to compare the effects of revegetation options. In Michigan,
Braden and others (1989) modeled the combined effects of
agricultural erosion and pesticides on fish populations by
using published chemical and sediment transport models and
habitat suitability indices.
Complementary effects usually involve interactions between
processes that modify each others’ rates. Statistical relations
between driving variables and responses can implicitly account
for consistent interactions, and studies that use regional
hydraulic geometries to predict the effects of altered flow on
channel geometry employ this approach (e.g., Jackson and
Van Haveren 1987, Rango 1970). More complex interactions
are better described using process-based computer models,
and all of the cited examples that evaluated progressive and
coexisting activities employed models.
Cascading and Interdependent Influences
No studies were found that predicted the effects of cascading
or interdependent influences. Some of the previously de-
scribed studies clearly reflect cascading influences: agricul-
tural development and urbanization usually are the reasons
that dams are built, clouds seeded, and flows diverted. How-
ever, each of these immediate causes was evaluated without
reference to the underlying land-use change that provoked
the activity. This approach is adequate for evaluating single
projects, but a full prediction of impact duration and future
changes would have to take into account the patterns of
change in the driving land use. For example, the cumulative
effects of a flow diversion designed to service a short-term
mining effort are quite different from those of a diversion
that supplies water to a rapidly growing city. Prediction of
CWEs from cascading influences will require coupling of
socio-economic models with physical and biological mod-
els. The socio-economic models would predict demand trends
in the underlying land use, while physical and biological
models would assess the effects of activities implemented to
satisfy the underlying demands.
Prediction of interdependent effects requires a clear
understanding of the underlying processes. Except in the
case of chemical interactions, interdependent effects are
rarely recognized.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Patterns of Predictive
Understanding
Published predictive studies addressed a single type of im-
pact or several closely related impacts. No studies were
found that provided a general overview of possible impacts,
although this task is often carried out in a planning context.
Each study isolated particular processes or mechanisms of
change and evaluated impacts in the context of those pro-
cesses, and none depended on analogy to paired-watershed
results. This is probably because the conditions driving the
change were unprecedented (e.g., ozone depletion), or the
unique nature of the setting prevented a suitable analog from
being found.
Studies were more likely to use computer modeling to
evaluate process interactions as the complexity of interactions
increased. Modeling results were usually used as indicators
of qualitative response and expected orders of magnitude,
and quantitative results were rarely accepted without
qualification. Most authors stressed that results were
preliminary and recommended additional basic research to
improve the models.
In contrast to retrospective studies, many predictions were
carried from the physical system to the biological because
the studies were often motivated by concern for a particular
biological resource. In other cases, the nature of physical
changes was predicted, but biological interpretation was left
to those interested in the biological resources. This approach
to biological impacts is a common one among physical
scientists for several reasons: biological responses are often
complex; interpretation usually requires a biologist; and a
physical change can influence many biological elements.
General Methods for Predicting
Cumulative Watershed Effects
Because of the importance of the CWE problem and the
mandate to predict CWEs, several researchers and land man-
agers have developed methods for assessing potential CWEs.
These procedures are usually designed for a particular land
use in a particular area, and they predict responses of a
particular target resource. Many produce disturbance indi-
ces that can be used to compare sites or management alterna-
tives, and most are in the form of a series of steps that can be
used reproducibly by resource professionals and technicians.
Although the methods’ limitations are recognized by their
developers, demands for ready-to-use estimators often propel
a model into general use, regardless of whether its assumptions
are appropriate at other project sites or whether it has been
independently tested or validated. Model results are occasionally
interpreted as a measure of likelihood for all possible impacts,
rather than for those which the model was developed to assess.
27
Examples of General Predictive
Methods
The most common CWE evaluation procedures used in Cali-
fornia are reviewed here, as are several models developed
elsewhere. In addition, a few procedures have been pub-
lished as general tools for evaluating particular aspects of
CWEs, and these, too, are reviewed. Methods are compared
in table 5.
Equivalent Clearcut Area (ECA)
One of the earliest CWE analysis procedures was developed
by the USDA Forest Service for use in northern Idaho and
Montana (USFS 1974, Galbraith 1975). The primary impact
of concern was channel disruption, and this was assumed to
be caused primarily by increased peak flows from reduced
transpiration due to logging. Channel disruption was as-
sumed to be an index of impacts to many beneficial uses, so
specific impacts were not considered.
Application of the model first requires calibration for an
area. The extent to which each management activity increases
water yield is determined as a function of vegetation type,
elevation, and age of the activity. Although these relationships
could be defined for many land uses, only those related to
timber management are usually included. Values for each
land type and use category are then compared to values for a
clearcut to calculate the area of clearcut that would produce
the same change, and this is used to calculate the equivalent
clearcut area (ECA) coefficients for the category. The amount
of monitoring data required for full calibration of model
coefficients is usually prohibitive, so professional judgment
is often used to define ECA coefficients.
Once the model is calibrated, application to particular
sites requires measurement of the area of each land-use
activity in each elevation zone and vegetation type. Areas
are multiplied by ECA coefficients and summed to calculate
total change in water yield, and altered water yield is assumed
to be proportional to altered peak flows. Allowable thresholds
for flow modifications are specified by law in northern
Idaho, and calculated values are compared to the mandated
thresholds. Allowable increases may be modified according
to the perceived stability of channels in an area.
The ECA model is not presented as a complete CWE analysis.
Provisions are not made to evaluate the effects of other types of
land use; other mechanisms of channel destabilization or peak-
flow increase are not analyzed; other types of environmental
changes are not considered; and specific impacts are not addressed.
In effect, the estimated increase in water yield is used as an
index of potential impact rather than as a predictor of impacts.
Because ECAs are calculated for a particular time, they
do not account for past impacts that might interact with
conditions at the evaluation time. Thus, the persistent effects
of old landslides are not accounted for in an ECA analysis.
Potential impact is assumed to be proportional to a year’s
transgression, and the recovery period for the impacted
resources is implicitly assumed to be the same as that for
28
water yield on a clearcut. This means that the model does not
apply to morphological changes that are cumulative through
time. Figure 4 illustrates this problem. In this case, the
driving variable (e.g., increased runoff, sediment input, or
similar) has a relatively quick recovery period, but the
impacted feature (e.g., channel width, volume of stored
sediment, smolt mortality, or similar) takes considerably
longer to recover from the effects of the temporary alteration
in the driving variable (fig. 4a). Even though the sequence of
land-use activities is carried out in such a way that the
driving variable has ample time to recover, the disturbance
frequency is too high to allow recovery of the impacted
feature between disturbances, and the impact accumulates
through time (fig. 4b). To assess a temporally cumulative
impact, recovery periods both of impacts and driving variables
would have to be considered.
A model can be applied to new sites only if its assumptions
are valid there. The ECA model assumes that (1) channel
disruption is caused by increased peak flows, (2) increased
peak flows are proportional to increased water yield, and (3)
increased water yield is proportional to area logged. If these
are valid for a particular area, then the model may be
appropriate, but assumptions must be tested carefully if the
model is to be applied with confidence. Several studies have
compared water-yield increases predicted by ECA with
measured changes. King (1989) showed a 44 percent
underestimate by ECA in basins smaller than those the model
A. EFFECT OF ISOLATED ACTIVITY
response of driving variable
Response
response of impacted feature
threshold of
concern (T.O.C.)
Time
B. EFFECT OF REPEATED ACTIVITY
response of driving variable
cumulative response of impacted feature
Response
T.O.C.
Time
Figure 4—Cumulative effect of differing recovery times for a driving
variable and an impact.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Table 5—General procedures for predicting CWEs

Equivalent Klock Watershed Equivalent R-1/R-4 CDF Questionnaire WRENSS Limiting Factor Rational Approach
Clearcut Area Cumulative Effects Roaded Area Sediment-Fish Analysis (Grant)
Analysis Model

Model type empirical association weighted indices use intensity index empirical process-base checklist physical process sum impacts; process physical process
Area applied northern Idaho, Mont. east Wash. Cascades Calif. Idaho Batholith California general western Wash., Oreg. west Oreg. Cascades
Can be used with calibration weighting method not with calibration only if relations with redesign to suit yes only if relations are yes
elsewhere? explained measured for locale local issues measured for region
Tested? Belt 1980, King 1989 no no not independently no some modules in progress no
Basin size third to fifth order < 4000 ha 200 - 20,000 ha type B,C streams 3rd, 4th order streams by site coho-bearing streams alluvial reaches
Result water yield, peak flow index index rates and populations qualitative various process rates smolt production critical flow
Impact vegetation change sheetwash, slides impacts increase with eroded sediment in all may be considered depends on module habitat change increased bedload
mechanism alters water yield and introduce sediment and increasing intensity channels impact fish alters fish survival transport alters

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

destabilizes channels destabilize channels of use and population channel
Land use logging, roads logging, roads logging, roads logging, roads logging, roads logging, roads not specified not specified
Impacts not specified not specified not specified fish water quality, fish, not specified coho smolt production not specified
recreation, others
Threshold yes, modified by yes yes; modified by site compared to natural no not applicable yes: pct of potential yes: channel-specific
identified? channel stability and impact type smolt production transport threshold
Calibration water yield increase slide survey, runoff impact intensity vs. relations between user must thoroughly depends on module relations between no
required by habitat, practice, relative to roads index; pct. change for sediment, habitat, and understand processes fish density and
elevation, region. uses relative to roads fish and impacts of region habitat parameters
Data required use history, area of use history, area of use history, area of use history, use unit may use any much data required population and habitat measured channel
for use each use unit each use use units, site factors areas, habitat and fish information source for most procedures inventories characteristics
surveys
Reproducible? yes; some judgment yes yes; some judgment; yes; some judgment diagnosis relies yes yes; some judgment yes
varies by district on judgment
Expertise:
calibration research or expert research or expert research or expert research not applicable research or expert research not applicable
operation low trained resource specialist resource specialists Reg. Prof. Forester resource specialist fisheries biologists expert
Comments does not handle does not handle intended as initial based on extensive analysis procedure collection of from point of view of addresses sensitivity
temporal CWEs temporal CWEs screening research in area not specified procedures impacted resource to change

29
was designed for, and Belt (1980; quoted in King 1989)
found a 38 percent underestimate in appropriately sized basins.
The theoretical foundation for the ECA method is weak.
Logging is known to increase water yield by reducing
transpiration, but this increase occurs primarily during the
drier seasons and rarely affects the highest peak flows.
However, peak flows may be significantly increased by
logging in areas subject to rain-on-snow events, because
snow accumulates more and melts faster in cleared areas.
This is likely to have a more significant effect on channel-
modifying peaks than altered transpiration. An index of clearcut
area may fortuitously address both mechanisms of change,
but numerical predictions are likely to be unfounded because
the underlying processes are different.
Klock Watershed Cumulative Effects Analysis
(KWCEA)
Klock (1985) adapted and combined elements from other
CWE assessment methods into an equation to analyze CWE
potential from timber management in the Washington Cas-
cades. The impact of concern is again channel destabiliza-
tion, but the driving mechanism is here assumed to be in-
creased sediment input. The KWCEA equation combines
factors for local climate (the R-factor of the Universal Soil
Loss Equation, Wischmeier and Smith 1978), susceptibility
to surface erosion (the USLE K-factor adjusted by a distur-
bance index for logging practice and recovery through time),
a landslide occurrence factor (a local landslide frequency
adjusted by site recovery factors), a topographic factor that
incorporates gradient and distance from a stream, and a
hydrologic sensitivity factor that indexes increased evapo-
transpiration after logging. Model application thus requires
information on soil characteristics, topography, landslide
frequencies associated with logged sites and roads, land-use
history, and areas of land-use types. Index values are calcu-
lated for each type of site in a watershed, normalized by
watershed area, and summed to provide an index of potential
impact. An index greater than 1.0 indicates a greater than 50
percent chance of “increased impact on the downstream
aquatic ecosystem.”
Unfortunately, no documentation was provided for the
derivation of model components, and procedures for assigning
relative weightings were not explained. Disturbance and
recovery coefficients presumably reflect measurements from
central Washington, so the equation must be recalibrated for
regions where runoff and erosion processes interact with
channels in different ways. But adjustment of component
weights for new areas is an indeterminate problem because
component indices do not represent physical quantities: there
are too many knobs to twiddle and too little information to
guide their twiddling. Klock (1985) did not indicate how
impact probabilities were related to index values, the types
of impacts considered, or if and how the model was tested.
Other references to its application were not found.
The KWCEA model assumes a relatively narrow view of
CWEs. Only the effects of timber management and roads are
30
considered, and the model is relevant only to those impacts
that might be generated by altered sediment input. Because
site-recovery functions are used to calculate each year’s
index, effects that accumulate through time are not addressed.
This oversight is particularly relevant to sediment-related
impacts, because introduced sediment may be stored for
long periods and cause long-term lag effects.
Equivalent Roaded Area (ERA)
USDA Forest Service Region 5 staff developed the ERA
method to index CWE potential from timber management
and roads. The original model assumed that channel destabi-
lization is the impact of most concern in California, and that
destabilization occurs primarily from increased peak flows
due to soil compaction.
The model has changed considerably since its initial release.
The recent version (USFS 1988, Cobourn 1989) extends the
procedure to address downstream impacts generated by several
mechanisms. Impact potential is indexed by relating the impacts
expected from each activity to that expected from roads. The
sum of indices for a watershed represents the percentage of
basin in road surface that would produce the same effects as
the existing or planned distribution of management activities.
Indices for different planning options can then be compared
to rank their potential for producing impacts.
Application of the method first requires identification of
important downstream values and the criteria necessary to
protect each. A use history is developed for the watershed,
sensitive sites are identified, and ERAs are calculated for
each activity with respect to the mechanism thought to be of
most concern. Values are summed for the watershed to
calculate the total ERA, and this is compared to the allowable
threshold identified for the area. If the calculated ERA value
is high, then the area is singled out for further evaluation by
other means.
Each National Forest is expected to identify focal concerns
and mechanisms and calibrate coefficients for characteristic
site types, and Haskins (1987) outlined this procedure for
the Shasta-Trinity National Forest. Recent practices often
cause less impact than older ones, so changes in land
management procedures must be identified and accounted
for, and recovery curves for various activities and site types
also must be defined. Calibration ideally incorporates
monitoring data for the identified impact mechanism from
different land use activities in that area, but the necessary
data are rarely available, and calibration usually depends on
professional judgment. Uses other than timber management
and road use can also be evaluated if their effects are
appropriately calibrated. Thresholds of concern (TOCs) are
to be identified independently for each area and are to take
into account inherent differences in sensitivity to impacts.
TOCs are usually defined by calculating ERAs for areas
showing different levels of impact. Users are expected to
exercise judgment in modifying ERA coefficients and TOCs
for particular sites.
The ERA method is essentially an accounting procedure
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
for assessing the instantaneous influence of past, present,
and planned activities on the potential for environmental
change. The method is designed to provide a screening tool
for identifying areas of particularly intense use rather than to
predict effects, but as an index of use intensity, ERAs are
likely to be grossly correlated with many types of impacts.
The method also provides a framework for organizing local
information on land-use impacts and mechanisms of change.
ERAs are not comparable between areas because the
method is customized to address issues relevant to each
implementation area. ERA coefficients defined for an area
where landsliding is the major impact have little relation to
those defined where the major concern is increased peak-
flows from rain-on-snow events.
Management activities are usually planned to maintain a
watershed’s ERA below an identified TOC. If the threshold is
approached or exceeded, then activities are reviewed to
determine whether they should be modified or delayed, or
whether existing conditions can be improved to lower the
ERA values. A basin is assumed to be healthy again as soon as
subthreshold ERA values are reattained, but the recovery times
actually required by impacted resources are not considered.
The method thus cannot be used to identify sites where impacts
persist from earlier disturbances, so the method is incapable of
addressing temporally accumulating effects: if recovery time
for the impacted resource is longer than that for the driving
variable, then impacts can continue to accumulate even though
the driving variable does not (fig. 4). Complementary effects
are also excluded, because the method requires identification
of a single impact mechanism. Monitoring programs are not
yet in place to assess the success of the most recent ERA
method in avoiding impacts, and earlier formulations of this
method were not independently validated.
The early formulations of the ERA model used the results
of a study in southern Oregon that showed peak flow increases
in a basin with 12 percent roaded area (Harr and others
1975) as the basis for identifying both the driving mechanism
for change (increased peak flows) and the TOC (12 percent
compacted area). However, these results are not transferable
to California’s geology and climate. Ziemer (1981), for
example, demonstrated no significant change in peak flows
in a California Coast Range watershed with 15 percent of its
surface area compacted. In addition, because channels respond
as readily to altered sediment inputs as to altered flows,
selection of increased peak flows as the single driving
mechanism does not fully address the problem. The most
recent implementations of the ERA model avoid these
problems by using ERA values primarily as an index of
land-use intensity, so the hydrological basis of the original
method is no longer important.
R-1/R-4 Sediment-Fish Model
USDA Forest Service researchers of the Intermountain Re-
search Station, resource specialists and managers from For-
est Service Regions 1 and 4, and university researchers
worked together to develop a method for predicting fish
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
survival as a function of sediment input in the Idaho Batholith.
The procedure has two parts: first, sediment yields are esti-
mated using a method described by Cline and others (1981),
and then the effects of increased yield on fish are calculated
using relations presented by Stowell and others (1983). The
R-1/R-4 model addresses settings where deposition of fine
sediment is the major impact on fish populations, and where
sediment is eroded primarily by surface erosion on logged
sites and roads. Components of the method are continually
revised to incorporate new research results, and model coef-
ficients are calibrated locally for each application area. De-
velopers of the model stress that it is directly applicable only
to the Idaho Batholith, and that results should be taken as
broad estimates of trends and relative impacts rather than as
precise predictions of change.
Sediment yield is predicted using relations developed from
extensive research on erosion process rates as a function of land
use in the Idaho Batholith. Average loss rates were calculated
for roads, logged areas, and burned sites, and these are applied
to the areas of each activity in a watershed to calculate on-site
loss rates. A sediment delivery relation from WRENSS (Water
Resources Evaluation of Non-point Silvicultural Sources, see
following section; USFS 1980) is used to calculate input to
streams, and a relation defined by Roehl (1962) allows estimation
of delivery to critical stream reaches.
Gravel siltation, summer rearing habitat, and winter
carrying capacity were identified as factors limiting salmonid
survival. Results of studies in the Idaho Batholith were
compiled to provide empirical correlations between substrate
embeddedness and sediment yield, between habitat measures
and substrate embeddedness, and between fish response and
habitat quality. To apply the model, calculations are carried
out for critical reaches downstream of a project area. If
natural siltation levels are known and conditions prior to the
project are measured, then the incremental effect of a planned
project can be estimated.
As presented, the model applies only to the Idaho Batholith
because its relationships depend strongly on runoff mode,
runoff timing, climate, sediment character, erosion process,
channel geometry, and species of fish. Use of the model in other
areas requires remeasurement of each of the component relations.
This procedure is one of the few that relate land-use
activities directly to a resource response, and is unique in its
recognition that impact recovery rates cannot be indexed by
recovery rates of the driving land use. Once a channel reach
is modified, the new condition becomes the baseline for
further changes, and effects that accumulate through time
can therefore be predicted. The R-1/R-4 model is not a
complete model for CWE evaluation because it addresses
only one type of CWE from one mechanism, but it is well-
founded on research results within the area for which it was
developed.
California Department of Forestry Questionnaire
The California Department of Forestry and Fire Protection
(CDF) has developed a procedure for use by Registered Pro-
31
fessional Foresters to assess CWE potential from timber man-
agement (CDF 1991). This procedure differs from all others
described in that it relies almost completely on the users’
qualitative observations and professional judgment, and it
provides only qualitative results. It also addresses a wider
variety of uses and impacts than other procedures, and in-
cludes many that are not related to water quality, such as
recreational, esthetic, biological, cultural, and traffic uses and
values. It was designed to be used within the time and access
constraints of timber harvest plan (THP) development, and a
nonquantitative approach was adopted to avoid the compla-
cency that accompanies a numeric, “right” answer.
Application of the method requires four steps for each
potentially impacted non-water-quality resource. The user is
first asked to identify significant beneficial uses in the
assessment area, and is then asked whether the planned
timber operation is likely to produce significant effects on
each of those uses. The third step requires the same evaluation
for the proposed project in combination with other planned
activities in the analysis area. Finally, the user is asked
whether significant cumulative effects are likely from the
proposed operation.
Analysis follows a different procedure for water-quality
resources. The analysis area for watershed assessments is an
“area of manageable size relative to the THP (usually an
order 3 or 4 watershed)” (CDF 1991). Existing channel
conditions are first inventoried, and adverse impacts from
past projects are identified. The user is then asked to judge
the likelihood of a variety of potential effects from the
proposed project, from expected future projects, and from
combined past, present, and future projects. “Past projects”
are generally restricted to those which have occurred over
the preceding 10 years, unless there is special knowledge of
an older “open sore,” and “future projects” are those which
are planned by the THP submitter over the next 5 years, or
which have been publicly announced by other users.
Likelihood is rated as high, moderate, or low, and further
analysis by specialists is recommended if CWE potential is
identified as high or moderate.
The CDF procedure is essentially a checklist to ensure
that the important issues have been considered. Although it
provides descriptions of possible CWEs, it does not specify
how the likelihood of their occurrence is to be evaluated, and
instead relies on the user’s ability to make qualitative
predictions based on observations of earlier projects in the
area. Validity of the results rests completely on the user’s
expertise, experience, and professional judgment, so results
are not necessarily reproducible. The specified spatial and
temporal evaluation scales are based primarily on feasibility
of evaluation, rather than on the nature of the potential
CWEs. This implicitly restricts the types of CWEs that can
be evaluated, although the procedure permits a larger scope
where it is deemed by the user to be necessary.
The procedure’s major strengths lie in its flexibility. This
is the only CWE evaluation method that requires assessment
of more than one type of impact from more than one type of
32
mechanism. It is also one of the few procedures that allow
evaluation of temporally accumulating impacts.
Water Resources Evaluation of Non-point
Silvicultural Sources (WRENSS)
The most complete process-based approach to evaluating
timber-management impacts is the series of procedures re-
ferred to as WRENSS (USFS 1980). This collection presents
quantitative evaluation procedures for a variety of water
quality impacts, including altered flows, sediment, and tem-
perature. Pollution by nutrients and chemicals is addressed
qualitatively, as are changes in dissolved oxygen. Although
not intended specifically to address CWEs, WRENSS meth-
ods are applicable to CWE evaluations.
Application of WRENSS to a CWE analysis would require
identification of likely environmental changes generated by
a project, likely downstream impacts, and the mechanisms
generating them. Appropriate WRENSS procedures would
be selected to estimate the magnitude of expected impacts
from a planned project, and these values would be added to
those calculated for existing projects to estimate the total
effect. Because the original focus for WRENSS was water
quality, the procedures do not address impacts on other
resources, and only the effects of timber management and
roads are considered. Evaluation procedures are independent
of one another, so modules can be replaced by improved
methods as they become available.
The WRENSS procedure for evaluating hydrological
change is based on computer simulation modeling of water
budgets. The program PROSPER (Goldstein and others 1974)
was used to develop relationships for rain dominated areas,
and WATBAL (Leaf and Brink 1973a, 1973b) for areas with
snow. Results are presented as graphs and tables that allow
users to estimate changes in evapotranspiration, flow duration,
and soil moisture for various logging plans. Stream
temperature changes are assessed using the Brown model
(Brown 1970).
Sediment modules include methods for estimating surface
erosion, ditch erosion, landsliding, earthflow activity, sediment
yield, and channel stability. Surface erosion is calculated
using a modified Universal Soil Loss Equation (Wischmeier
and Smith 1978) and sediment delivery relations, and ditch
erosion is assessed by calculating permissible velocity. The
landslide module is a step-by-step guide to performing a
landslide inventory for an area: the area is subdivided into
uniform subareas, hazard indices are calculated for each,
and slides are inventoried in representative areas to determine
characteristic volumes and delivery ratios as a function of
hazard index and land use. Sediment yields are estimated
using results of monitoring and process evaluations, and
channel stability is indexed by changes in sediment and
flow.
WRENSS is neither a CWE model nor an evaluation
procedure, but is simply a collection of tools useful for
impact evaluation. A CWE analysis using WRENSS would
be flexible enough to handle a variety of impact mechanisms,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
but it would need to use additional methods to assess the remain stable. Grant presented equations to calculate flow
effects of other land uses and to evaluate impacts on particular thresholds for sediment transport as a function of particle
resources. Implementation of most of the procedures requires size and channel geometry. Projected flow increases can then
training in hydrology or geomorphology, and calculations be compared to threshold discharges to evaluate their potential
are often complex and time-consuming. WRENSS is the for altering a channel. The Rational Approach can also provide
only method described here that is capable of estimating the an index of channel stability.
magnitudes of different types of watershed changes. The Rational Approach illustrates the potential use of
Each procedure presented in WRENSS was developed by process-based information for quantifying the potential effects
researchers and resource specialists with relevant expertise, of changing conditions, but it was not intended to provide a
but procedures vary widely in sophistication, approach, and full CWE analysis. Use of the method to evaluate CWEs
accuracy, and many have been superseded by more effective would require its coupling with models that predict flow
methods. Some of the methods have been intensively tested, changes from land use, and with methods to analyze channel
while others have not been validated at all. destabilization by other mechanisms, such as increased
sediment input or riparian disruption.
Limiting Factor Analysis (LFA)
Reeves and others (1989) developed the LFA as a procedure
for identifying factors that limit coho smolt production in
Discussion of General CWE
coastal Oregon and Washington, and the procedure can be Assessment Procedures
adapted for use in CWE analysis. Unlike the approaches
The approaches discussed above fall into several categories
described above, LFA is designed from the point of view of
and are summarized in table 6. The ERA, KWCEA, and
a particular impacted resource. Any procedure designed to
ECA methods produce indices that represent the potential
predict CWEs, rather than merely to assess their likelihood,
for CWE generation in a watershed, while the CDF ques-
must incorporate a component such as this.
tionnaire provides a qualitative assessment of CWE likeli-
The LFA model is in the form of a dichotomous key that
hood. The other techniques produce quantitative measures
leads users through computations to estimate smolt production
of rates, volumes, or populations. The R-1/R-4 model quan-
from data on physical habitat in a watershed. Model application
tifies a limited cause/effect relation, and can be used to
requires detailed surveys of fish populations and areas
assess a particular type of effect in areas for which the
occupied by various habitat types in a watershed. The
relation is defined. The Rational Approach (Grant) and the
procedure is based on extensive fish and habitat surveys that
LFA evaluate effects from the point of view of the impacted
were compiled to disclose patterns of habitat use in the
resource without reference to the land uses triggering the
Pacific Northwest. LFA can thus be applied only to sites
impacts. Finally, WRENSS is a collection of procedures for
with population-habitat relationships characteristic of the
evaluating several types of impacts.
development area, but the model could be recalibrated for
ECA, ERA, KWCEA, R-1/R-4, and CDF methods were
other areas if limiting factors are the same and appropriate
all designed to analyze CWEs. The first four methods assume
relationships can be measured.
that CWEs occur primarily by channel disruption or
Because LFA sums smolt production from different habitat
sedimentation, and changes affecting other mechanisms are
categories, the impacts of changes in habitat distribution can
not addressed. Models may be useful where this assumption
be calculated. Relations between cumulative habitat change
is met, but none of these methods can address all CWEs
and land-use activities would also be required for prediction
possible in an area. None of the four can be used to evaluate
of CWEs, and these are produced in some form by most
on-site or complementary-influence effects, and only the R-
other models described here. A procedure capable of assessing
1/R-4 method is capable of addressing effects that accumulate
impacts of timber management on fish would need to couple
through time. The CDF method requires evaluation of a
a trigger-based model such as ECA or ERA with an impact-
broader range of mechanisms and impacts and allows
based model like LFA. The model is currently being validated
evaluation of temporally accumulating effects, but analysis
by the Oregon Department of Fish and Wildlife, but results
procedures are left to the professional judgment of the user,
of the study are not yet available.
and the recommended spatial and temporal scales for analysis
are too small to address some types of CWEs. Because of
Rational Approach (Grant)
these limitations, none of the approaches provides a complete
Grant (1987) described a process-based procedure for estab-
evaluation of CWEs.
lishing thresholds of concern for flow-related channel disrup-
The scientific basis for most of the general methods is
tion. As was the case with LFA, this method is designed from
poor. The ECA method is based on extensive data showing
the point of view of the impact rather than the impacting use.
that decreased vegetation cover augments water yield by
Grant reasoned that channels can be disrupted only if
decreasing evapotranspiration losses. However, the method
their beds and banks are remolded, and this requires
then assumes that changes in peak discharges are proportional
mobilization of bed sediment. If increased flows do not
to changes in water yield, a relation contradicted by most
affect the frequency of sediment transport, then a river will

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 33

 Table 6—Attributes of CWE evaluation procedures

ECA KWCEA ERA S/F CDF WRENSS LFA RA

Uses considered
Timber management x x x x x x . .
Roads x x x x x x . .
Mining c . c c x (x) . .
Agriculture c . c c x (x) . .
Recreation c . c c x (x) . .
Urbanization c . c c x (x) . .
Impoundments . . . . x . . .
Mechanisms:
Runoff change x . x . x x x .
Peak flow change x . x . x x x x
Sediment change . x x x x x x .
Organic debris change . . . . x . x .
Nutrient change . . . . x x . .
Input of toxics . . . . x x . .
Temperature change . . . . x x x .
Impact:
Channel stability x x x . x x . x
Fish . . . x x . x .
Recreation . . . . x . . .
Water quality . . x x x x . .
Sedimentation . . x x x x . .
Form of result:
Index . x x . . . . .
Rate or value x . . x . x x x
Qualitative . . . . x x . .
Scientific basis:
Theoretical . . . x n (x) x x
Empirical x n x x n (x) x n
Validation . n . . . (x) * n
Source:
Research . n . x . x x x
Management x n x x x . x .

x Model applies ECA Equivalent Clearcut Area

. Model does not apply KWCEA Klock Watershed Cumulative Effects Analysis
c Model could apply if calibrated ERA Equivalent Roaded Area
n Not specified or unknown S/F R-1/R-4 Sediment Fish Model
(x) Varies with procedure CDF California Department of Forestry
* In progress Questionnaire
WRENSS Water Resources Evaluation of Non-point
Silvicultural Sources
LFA Limiting Factor Analysis
RA Rational Approach (Grant)

research. Too little information is provided about KWCEA
to judge its theoretical validity. The ERA method originally
incorporated the premise that peak flows would increase if
12 percent of a watershed were occupied by roads, as suggested
by preliminary results of an Oregon study (Harr and others
1975). These results have little relevance to other regions,
and subsequent work showed that no threshold existed even
in the study area (Harr 1987). Although the premise has
since been dropped and the conceptual background expanded,
the ERA model retains the framework imposed by this
assumption. In contrast, the R-1/R-4 model is based on
research results from the area for which it was developed. Its
assumptions are clearly explained, the data on which relations
are based are presented in the procedural guide (Stowell and
others 1983), and research continues to improve the model.
WRENSS, LFA, and the Rational Approach (Grant) were
not developed specifically as methods for CWE analysis, but
can be used as components in evaluations. All are well-
founded in basic research. The relations employed by the
LFA represent decades of field measurements, the Rational
Approach (Grant) is based on fundamental sediment transport
mechanics, and many of the methods employed by WRENSS
are founded on extensive experiments and large data bases.
Other WRENSS techniques, however, are less well supported.
Surprisingly, none of the general methods (ECA, KWCEA,
ERA, CDF, R-1/R-4) have been adequately validated by
documented independent studies. Only the ECA method has
been tested in published reports, and monitoring results
demonstrated a 38 to 44 percent underestimate in predicted
water yield (Belt 1980, King 1989). Methods that produce

34 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

index values or qualitative results cannot be tested by
comparing predictions with measured values, and validation
requires an audit of past predictions and comparison of
measured impacts with calculated indices. Because most
data required by the ECA, ERA, KWCEA, and R-1/R-4
methods can be reconstructed from photographs and maps,
watersheds can be evaluated for conditions pertaining in the
past. Retroactive predictions of impact potential could thus
be compared with present evidence of impacts to permit
testing of model validity without lengthy monitoring. Standard
records of impacts associated with evaluated projects would
be extremely useful for model validation and comparisons.
Bailey and Hobbs (1990) suggested a framework for auditing
the effectiveness of Environmental Impact Assessments. They
recommended establishing a database that includes the type
of project, environmental conditions, predicted impacts,
methods used for prediction, and observed impacts.
Table 5 and table 6 reveal a pattern in model characteristics.
When methods originate from management agencies, they
tend to be simple, incomplete, theoretically unsound,
unvalidated, implementable by field personnel, and heavily
used. Methods developed by researchers are more likely to be
complex, incomplete, theoretically sound, validated, require
expert operators, and not used. Only in the cases of the R-1/R-
4 method and the LFA have management and research expertise
been combined to design the methods, and these rank among
the highest in defensibility, scientific basis, and utility.
Methods Under Development
Three additional general CWE evaluation methods have
taken form since 1990. All are currently under development
and are not yet tested, but prototypes of each are expected to
become operational during 1992.
Sequoia National Forest Method
In response to litigation, resource specialists of the Sequoia
National Forest are currently developing a CWE evaluation
procedure that is intended to relate the ERA sediment-pro-
duction index directly to the magnitude of environmental
change. The method is innovative in its proposed goal of
addressing the full range of beneficial uses that might poten-
tially be impacted and in directly incorporating a program for
monitoring the accuracy of predictions. However, the method,
as currently planned, shares with the standard ERA approach
the problems of: 1) not evaluating impacts of land uses
coexisting with timber management, such as mining; 2) not
permitting assessment of temporal effects because only the
recovery times of impacting uses are to be considered; and 3)
considering only a single mechanism for generating impacts.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
TFW Method
The Timber/Fisheries/Wildlife (TFW) group was established
in response to intensifying conflicts over land management
and impacts in Washington State. Composed of representa-
tives from State agencies, tribes, timber industry, and envi-
ronmental organizations, the group is dedicated to resolving
timber-management-related environmental conflicts through
a process of consensus and cooperation. A primary goal is
development of a method for evaluating CWEs. Current
plans call for a hierarchical approach to evaluation: water-
sheds will first be screened by interdisciplinary teams of
resource specialists to qualitatively assess their sensitivity to
environmental change, and if basins are found to be sensi-
tive, they will be scrutinized by the appropriate experts to
define more precisely the potential impacts and management
alternatives. Guidelines for evaluation are now being de-
signed, and a research program is planned to provide new
evaluation methods where existing approaches are inad-
equate. As planned, the method will address primarily the
effects of timber management on fisheries resources, and so
it will not constitute a general CWE procedure. However, it
is an extremely important development in CWE evaluation
because of the close cooperation of disparate interest groups
in development of the method.
NCASI Method
The National Council of the Paper Industry for Air and
Stream Improvement (NCASI) is also developing a hierar-
chical approach to CWE evaluation for application on tim-
ber-industry lands throughout the five western forested states.
The initial screening procedure will encompass the full range
of impact mechanisms and beneficial uses present in the
region and is expected to be in the form of a computer-based
expert system that will lead nonspecialist users through a
diagnostic procedure. The expert system will also suggest
relevant evaluation procedures for the watershed, and these
will be carried out during the second phase by appropriate
specialists. Although similar in structure to the TFW ap-
proach, the NCASI method will address all relevant types of
downstream impacts, rather than being restricted to fisher-
ies-related effects. The plan also calls for monitoring the
method’s effectiveness. On-site CWEs will not be assessed,
however, and it is not yet clear how analysis will proceed in
areas of multiple ownership.
Prospects for a General CWE
Evaluation Procedure
A complete general procedure for assessing the potential for
CWEs does not yet exist, although specific cumulative im-
pacts have been routinely predicted for decades. The strengths
and weaknesses of existing assessment methods provide
useful guidance for developing a general method.
35
Attributes Required of a General Method
Comparison of existing CWE methods reveals several
attributes that a successful general method will need to
incorporate:
1. Recognition that CWEs may be generated by complexly
interacting mechanisms. Methods that assume a single
mechanism may work in limited areas, but cannot be applied
where processes, uses, and impacts are different.
2. An ability to evaluate effects from many types of land
use. Methods that address only timber-management-related
impacts will become less useful as other types of land use
increase in wildland watersheds and as the spatial scope
required of CWE analyses increases.
3. An ability to address the range of impacts likely at all
sites downstream. The potential for CWE-based litigation
does not stop at an arbitrary radius from an activity site, and
indices that reflect use intensity or condition of a resource
cannot represent susceptibility to all potential impacts.
4. Methods flexible enough to allow site-specific
prescriptions based on local conditions that influence impact
generation.
5. Evaluation of the effects of past impacts. Recovery
curves for triggering mechanisms describe persistence of the
triggers, but they provide no information about the persistence
of impacts. A site’s impact history may influence its
susceptibility to future impacts.
6. Use of the best available technology. Court tests for
adequate CWE evaluation may increasingly rest on whether
the best available technology was used, and this question
may decide the outcome of cases when opposing conclusions
are presented. Because our understanding of natural systems
is continually improving, no method will be definitive for
long, and a general CWE method must allow for modification
as analytical techniques improve.
7. Methods to track the spatial distribution of activities,
impacts, and land-use histories at each site. Geographic
information systems currently provide the best available
method for organizing and manipulating this type of spatial
information.
8. Verification by statistically sound comparisons of
predicted and observed impacts.
Methods that fit these criteria will require that users be
capable of applying sound professional judgment and have a
strong understanding of the local conditions that affect process
rates; implementation will thus need to be carried out by
specialists. A successful model would need to include multiple
modules in order to provide appropriate tools for each of the
variety of impacts likely to be encountered, but implementation
at a site would require that only a subset of those modules be
used. Appropriate modules would be selected according to
the land uses present, the mechanisms of change they
represent, and the resources they might affect. Common
patterns of cause and effect could be characterized regionally
to decrease the work necessary for specific applications.
36
Development and Use of a General Method
As will be described in Chapters 5, 6, and 7, methods
already exist for evaluating the effects of most types of
environmental change on watershed processes, and the
resulting effects on many beneficial uses can also be predicted.
These procedures could be compiled into a collection of
evaluative tools, and research could focus on improving
fundamental understanding of the weakest modules. As
understanding grows, teams of researchers and users could
translate research developments into improved tools to take
the place of preliminary modules. This continually updated
set of procedures would provide a general evaluation method
for CWEs.
Application of a general method would first require its
customization to address regional issues and conditions. The
selected modules would then be calibrated and verified for
the region, and application to specific sites would then become
relatively routine. A mechanism for monitoring the model’s
validity and utility must be incorporated into each regional
customization. MacDonald and others (1991) reviewed a
variety of monitoring techniques and provided criteria to aid
in selecting appropriate monitoring methods and strategies.
Once the general procedures are compiled, different levels
of expert input and time commitment would be required at
different stages of implementation. Regional customization
and application could proceed according to the following
framework:
I. Develop a regional submodel by selecting appropriate
modules
A. Use CDF-style analysis to identify:
1. Resources and values that might be impacted
2. Types of impacts likely to be of concern
3. Types of changes in watershed processes capable
of causing these impacts
4. Land-use activities present or planned in
watersheds in the area
5. Types of changes in watershed processes that
might occur from these uses
B. Identify impact mechanisms that will need to be
included in the submodel by comparing likely changes
to those that might cause impacts
C. Select modules that address identified regional needs
II. Develop database required for implementation (will
usually involve GIS)
A. Collect information on land-use history in the region
at whatever scale is necessary to meet the agency’s
goals
B. Collect information on environmental parameters
throughout the region (e.g., soil type, vegetation, etc.)
C. Determine tolerable impact levels for potentially
impacted resources using:
1. Published tolerance levels
2. Public input
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 3. Local studies
D. Calibrate values required by the selected modules,
using:
1. Published values
2. Local monitoring records
E. Identify types of data that can be collected during use
to improve the model
III. Test submodel
A. Verify each calibrated module
1. Carry out short-term monitoring studies
2. If possible, compare retroactively predicted
results
of each module with:
a. Existing long-term monitoring data
b. Surrogate time-series data (e.g., air photos)
B. If possible, compare retroactively predicted results of
entire submodel with:
1. Existing long-term monitoring data
2. Surrogate time-series data (e.g., air photos)
C. Establish monitoring sites for the types of changes
identified in IA2
1. To test individual modules
2. To test the entire submodel
3. To allow future improvement of calibrated
coefficients (as per IIE)
IV. Package submodel
Modules could include computer algorithms, questionnaires,
“cookbook” field procedures, predictive equations, and other
approaches; methods will vary with each type of mechanism
to be evaluated. Some may produce quantitative results,
some qualitative. Calibrated coefficients (IID) will need to
be incorporated into the modules, and instructions for using
the customized submodel and database will need to be
prepared. Users will need to be trained.
V. Apply submodel to specific projects
A. Evaluate CWEs from project
1. Examine the site and downstream resources to
ensure that the assumptions on which the
regional
submodel is based are valid for this
implementation.
2. Carry out evaluation procedures specified by the
submodel
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
3. Assess reasonableness of predictions by:
a. Examining nearby sites where similar projects
have been carried out
b. Comparing predicted values with published
results for similar sites
c. Comparing predictions with monitoring data
for the area
B. Enter procedures used, environmental conditions,
and predictions into a database to allow future
auditing of the submodel’s effectiveness
C. Establish a monitoring program to detect likely changes
1. Use predicted changes to identify sensitive
parameters
2. Measure status of parameters at the time of
analysis
3. Periodically assess changes in parameters after
the project is carried out
VI. Review submodel
A. Periodically assess success of submodel
1. Assess condition of resources below previously
evaluated sites
2. Use auditing database to compare observed
conditions with those predicted
3. Compare observed conditions with tolerance
limits established in IIC
4. If the submodel is not successful, identify and
reconstruct weak modules
B. Periodically update coefficients using data from
ongoing monitoring
C. Incorporate updated modules from the general model
as they become available
Extensive input from resource experts and researchers
would be required during the development phase, and lesser
inputs during implementation phases. Expert input is also
required to assess features of project sites or downstream
resources that might require modification of the submodel.
A continuing research component is necessary to update and
improve modules.
This approach to CWE evaluation can be implemented
using existing information and technology. The research
described in Chapters 5, 6, and 7 provides the basis for module
construction, but teams of researchers and users will be needed
to turn this information into useful management tools.
37
Chapter 5
Effects of Land-Use Activities on Environmental Parameters
L and-use activities can directly affect vegetation, soil
properties, and topography, and can import or remove
water, chemicals, pathogens, and fauna. Changes in these
environmental parameters, in turn, provoke changes in
watershed processes and affect other resources. CWEs are
generated when two activities influence the same
environmental parameter, transport process, or beneficial
use. On-site CWEs can result directly from altered
environmental parameters, whereas off-site CWEs reflect
changes in watershed transport processes. In both cases,
CWEs ultimately result from alteration of these few
environmental parameters.
Land use can directly influence environmental parameters,
and these changes can then provoke indirect changes in other
environmental characteristics. Conversion of vegetation to
increase runoff is a direct effect, but vegetation changes caused
by changes in runoff are indirect effects. Most direct effects
can be observed and measured, and many are predictable.
This section reviews direct and indirect effects of land use
activities on environmental parameters and describes studies
that evaluate these changes. Subsequent chapters discuss the
effects of environmental changes on watershed processes
(Chapter 6) and beneficial uses (Chapter 7). This organization
is adopted because many activities provoke similar
environmental changes, and impacts reflect the type of change
rather than the land use generating it. Thus, many activities
compact soils, but compaction from each source has the same
effect on the hydrologic system. The effect of specific land-
use activities on compaction is described in this chapter, and
the effect of compaction on hydrologic response is discussed
in the next. These chapters represent a brief overview of the
many disciplines involved, and provide only an introduction
to each topic. Recent publications and reviews are cited to
allow readers to pursue topics of particular interest. The
review focuses on activities related to timber management,
but this is rarely the only land use in a watershed. The effects
of other uses are also reviewed, because CWE assessments
must take into account all types of land use in a basin.
Some features are common to many land-use activities:
roads are an integral part of most land use, and water
development accompanies many uses. These features are
considered first because of their general relevance.
Roads
Most land use requires roads. Forest road construction and
design are usually controlled by county grading ordinances
or by state and federal regulations that mandate best man-
38
agement practices (BMPs). However, little is known of how
closely these regulations are observed or how well they
satisfy their intended goals. Recent studies showed that 17
percent of culverts at an Oregon site were too small to pass
the design flow (Piehl and others 1988b), and even appropri-
ately sized culverts were often constricted by debris, sedi-
ment, or structural damage (Piehl and others 1988a). Such
chronic under-design is likely to result in cumulative im-
pacts during storms that approach or exceed design stan-
dards. Roads also strongly influence other land uses. They
provide recreational access to remote areas, and their con-
struction requires development of quarries for aggregate and
surfacing gravel.
Roads and Vegetation
Roads cause appreciable vegetation changes. Trafficked sur-
faces are usually bare, and roadsides are often cleared to
increase visibility and aid maintenance. Exotic species intro-
duced to reduce erosion on roadcuts may spread to become
pest plants. Kudzu vine, for example, was once used to
control erosion and is now a major weed species in the
southeastern United States. Some seeds can be dispersed by
vehicles, and a few plants have distribution patterns that
correspond to road and railway corridors.
Other effects on vegetation are indirect. Timber
management and minerals exploration establish many roads
that are abandoned after use, but regrowth on compacted
soils is slow. Exposed mineral soils and oversteepened
roadcuts and fills provide unstable and nutrient-poor substrates
that inhibit revegetation, and de-icing salt suppresses
vegetation in areas receiving road drainage. Fleck and others
(1988) and Hofstra and Hall (1971) measured damage from
road salt to birch and pine, respectively, and demonstrated
decreased regeneration and increased foliar damage with
increased salt exposure. Walker and Everett (1987)
documented changes caused by road dust in tundra plant
communities. Species distribution changed to reflect the
increased mineral-soil content at the soil surface and the
shortened duration of snow cover caused by quicker melting
of dusty snow.
Clearing for road construction exposes new forest margins
that are susceptible to blowdown. Canopy openings along
roads can accelerate growth in adjacent trees, although rapid
drainage above roadcuts may suppress growth upslope (Pfister
1969). Disruption of soils and physical damage to trees can
increase their susceptibility to pathogens (Hansen 1978),
and pathogens can be carried by vehicles and distributed
along roadways. Spores of the fungus responsible for Port
Orford root disease can be dispersed by wheel traffic (Zobel
and others 1985).
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Roads and Soils
Traffic forms dust on dry soils or throws particles into
suspension if runoff is present, and soil textures on roads are
often modified by applying gravel to control erosion and
improve trafficability. Both paved and unpaved road sur-
faces are usually almost impermeable. Traffic collapses soil
pores on unsurfaced roads, and this compacts soils and alters
their erosivity (Voorhees and others 1979). Surfacing grav-
els are often cemented by fine sediments introduced by
gravel breakdown and by “pumping” of wet substrate into
gravels. Road construction disrupts soil profiles and often
strips roadsides of topsoil. Soil and rock excavated during
road construction are exposed to erosion on spoil heaps,
sidecast, and fill emplacements.
Roads and Topography
Road construction generally requires topographic changes
that alter drainage patterns and affect slope stability. Benches
are cut and fill is emplaced to support roads; embankments
are built to cross waterways; and revetments and levees are
constructed for protection from channel erosion and flood-
ing. To minimize the area impacted, roadcuts and fills are
often constructed at gradients near their stability limit. Ex-
cessive amounts of organic debris are occasionally incorpo-
rated into fills, which then become increasingly susceptible
to failure as the organics decompose.
Stream crossings and road drainage structures modify
channel morphology, and slope failures and gullying are
common where roads divert or restrict passage of channel
flows. Poorly designed drainage structures create knickpoints
and plunge pools along channels. The effects of these
topographic changes on water and sediment production and
transport are described in the following chapter.
Roads and Chemicals
Chemicals are introduced during road construction and use.
Oil products collect on road surfaces and are removed by
runoff. Where leaded fuel is still used, lead is concentrated in
roadside soil and vegetation (Motto and others 1970) and at
dump sites for plowed snow (Lockery and others 1983). Salt
used to de-ice roads accumulates where road drainage col-
lects (Hofstra and Smith 1984), and McBean and Al-Nassri
(1987) described the distribution of salt splashed from roads
by traffic. Hofstra and Smith (1984) found that salt accumu-
lation over a 20-year period decreased the availability of
calcium and magnesium to plants. Surfacing materials can
also modify the chemical composition of water draining from
roads, and Helvey and Kochenderfer (1987) demonstrated
increased pH for runoff from limestone-gravelled roads. Toxic
chemicals can be spilled during transport, and herbicides are
commonly used to control roadside shrubs. Bare roadcuts
and fill-slopes are often fertilized to promote revegetation.
Atmospheric pollutants disperse over wider areas than
those introduced on the ground. Recent decreases in
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
atmospheric and aquatic lead correspond to decreasing use
of leaded gasoline and demonstrate the past importance of
traffic as a source of lead (Alexander and Smith 1988,
Elsenrelch and others 1986). Traffic-related atmospheric
pollution can alter nutrient availability. Fenn and Dunn (1989)
showed that litter decomposition rates in the San Bernardino
Mountains of southern California increase with pollution
impact and may reflect an increase in litter nutrients from
premature foliage loss.
Impoundments and Water
Development
Water supplies are developed for domestic, industrial, and
agricultural use. Development usually requires transport and
storage of surficial runoff or groundwater, and hydrologic
processes may be altered to augment runoff. Clouds may be
seeded to increase precipitation, fences built to increase
snow accumulation, or vegetation reduced to decrease evapo-
transpiration. Reservoirs contribute not only to water sup-
ply, but also to power generation and flood control, and a
reservoir’s effects are partially determined by its purpose.
All impoundments alter immersed vegetation and that on
their shores. Seasonal fluctuations in reservoir levels prevent
establishment of typical riparian communities, and vegetation
within the range of fluctuation is usually limited to grasses
and sedges. Shallow reservoirs frequently support a high
concentration of algae, which contributes organic material
to downstream reaches.
Altered low flows and peak flows modify downstream
riparian vegetation. Increased variability of discharge usually
inhibits riparian vegetation, while more evenly distributed
flows allow riparian communities to encroach on channels
(Knighton 1988, Petts 1977). Decreased flooding promotes
colonization and establishment of mature plant communities
on floodplains. Harris and others (1987) documented changes
in riparian communities below dams, but noted that the
environmental characteristics of specific streams must be
known before response can be predicted. Watersheds from
which flow is extracted may show decreased vegetation covers,
and plant communities may change to reflect drier conditions.
In contrast, watersheds receiving transported water often
show higher cover densities and more mesic communities.
Construction of dams and diversions strips wide areas of
vegetation and topsoil, and sites used by construction and
maintenance vehicles are often heavily compacted. Stockponds
are very small impoundments, but they are common. In some
areas, stockponds provide the only surface water during the
dry season and may create riparian habitat where none
previously existed. These ponds become focal points for
livestock and wildlife use, and vegetation cover may be reduced
on adjacent land. Periodic inundation of soils on reservoir
margins alters their physical and chemical characteristics.
39
 Impoundments and diversions alter topography by creating
lakes or artificial drainage divides. Reservoirs may induce
landsliding by exposing steep hillslopes to wave action and
to increased pore pressures from raised water tables. Earthen
stockpond embankments are rarely protected from erosive
outflows, and breached dams are frequent in some areas.
Large reservoirs may increase topographical loading and
groundwater pore pressures enough to cause seismic activity,
and filling of one Indian reservoir apparently generated a
magnitude-6.5 earthquake (Haws and Reilly 1981). Most
dams are topographical barriers to aquatic animals.
Large amounts of oxygen are added to flow by whitewater
turbulence. Impoundments can remove this source, and Thene
and others (1989) constructed a model to predict the magnitude
of this impact. Oxygen content is also decreased by
decomposition of organic sediments trapped in reservoirs,
and anoxic water often collects at depth. Dominy (1973)
cited anoxic water as a hazard to fish, and also attributed
several fish kills in eastern Canada to nitrogen supersaturation
of reservoir outflows. Decomposition of immersed vegetation
can eutrophy new impoundments and modify nutrient cycling
and water chemistry.
Water yield can be increased by reducing evapotranspiration
losses in a watershed. Some forest lands have been converted
to grassland to increase runoff, and riparian vegetation is
occasionally removed for the same purpose. Mechanical
vegetation removal disrupts and compacts soil. Clearcutting
for water production produces the same responses as timber-
motivated cutting, and burning can induce hydrophobic soil
conditions. Use of herbicides to reduce evapotranspiration
and enhance water supplies can influence the chemical
environment, and chemically induced water repellency has
also been proposed to increase runoff.
Water developments often import water from other watersheds.
Transbasin inputs may locally increase groundwater and soil
moisture levels, but can severely decrease flows in source basins.
Where groundwater is pumped, heavy use may lower the regional
water table and cause salt-water intrusion or subsidence.
Timber-Management Activities
Timber management encompasses all activities associated
with growing and cutting trees, including yarding, site prepa-
ration, and planting. Road densities are high in logged areas,
and impacts from timber management are often overshad-
owed by those of roads. Road-related impacts are described
in an earlier section.
Timber Management and Vegetation
The dynamics of natural forest communities must be under-
stood to interpret the effects of vegetation changes induced
by timber management, but natural disturbance frequencies,
patterns, characteristics, and recovery rates have rarely been
40
determined. Natural frequencies and impacts of drought,
insect infestation, and blowdown are poorly understood, and
little is known of the susceptibility of different forest types
to infestations. A few studies have examined infestation
patterns within a forest community (e.g., Roe and Amman
1970). Fire frequencies have been determined at many sites
by dating scarred vegetation (e.g., Arno and Gruell 1983),
and Arno and Sneck (1977) described a procedure for such
studies, but data are still too few to reveal general patterns.
Hemstrom and Franklin (1982) compared fire disturbance
with other natural disturbances in Mt. Rainier National Park
and showed fires to be more significant than either lahars or
snow avalanches. Native Americans practiced vegetation
management by prescribed burning, but their strategies are
also poorly understood (Lewis 1973).
Ecological studies show that disturbance size, intensity,
and frequency influence species composition and community
structure (Oliver 1981). Lorimer (1989), for example,
documented differences in communities colonizing small
and large forest gaps. Parker (1986) found that lodgepole
pines dominate in several situations in the Sierra Nevada:
where disturbance is frequent, stands of this colonizing species
are even-aged and susceptible to successional replacement,
but stands are present at infrequently disturbed sites only
where site conditions are poor, and these stands are uneven-
aged and self-perpetuating. Response is also strongly
influenced by the types of species and communities present.
Halpern (1988) found that community resistance and resilience
in the Oregon Cascades is lowest for forest communities
with low species diversity.
The short-term effects of most timber-management
activities on vegetation are well known because of their
economic significance. Regeneration of marketable species
after logging has been widely researched, and many studies
correlate regeneration and growth rates to site indices.
Strothmann (1979) measured regeneration rates on clearcuts
in northwest California to relate stocking success to site
conditions and management practices, and showed that small,
tractor-yarded cuts planted in December, January, or May
were the most successful in the area examined. Less is
known of longer-term effects. Veblen and Lorenz (1986)
used evidence from 80-year-old logged sites in Colorado to
identify the effect of site characteristics on successional
patterns. On good sites, climax species dominated by
persevering through the successional stages, but at harsh
sites, a pioneering stage was necessary to make the
environment habitable for later-arriving climax species.
Abrams and Scott (1989) described Michigan forest
communities in which the commercial species are components
of an early successional stage, so that clearcutting accelerates
succession by releasing hardwood climax species.
Understory succession is less well understood. Everett
and Sharrow (1985) found that understory assemblages
remained after tree removal in a pinyon-juniper forest, but
species diversity declined and ground cover increased. Dyrness
(1973) documented differences in understory regeneration
rates and species composition on burned and unburned
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
clearcuts in the H.J. Andrews Experimental Forest in the
Oregon Cascades, and Halpern (1989) compared the long-
term distribution and growth patterns of residual and invading
species at that site and a nearby site.
The effects of yarding depend on the method used, with
physical disruption of ground-cover vegetation decreasing as
trafficked area decreases. High-lead and skyline methods are
less disruptive than tractors and skidders, and helicopter yarding,
yarding over a snow-pack, and balloon yarding further reduce
disturbance. Many studies document vegetation and litter
disturbance by different yarding methods (e.g., Dyrness 1965,
1972, Klock 1975, Miller and Sirois 1986, and Ruth 1967).
Sites are often manipulated after logging to promote natural
or plantation regrowth through burning, physical site
preparation, planting, fertilization, brush suppression, and
thinning. These activities affect vegetation communities by
removing vegetation, altering community composition, and
influencing growth rates. Lewis and others (1988) documented
undergrowth succession following several methods of site
preparation in combination with grazing in Florida. Most
communities are redesigned to promote a high-value species
and hardwoods are suppressed, so species diversity decreases.
Different methods of burning affect vegetation in different
ways. If slash is piled and burned, most disturbance is from
tractor work, but broadcast-burned sites are affected primarily
by the fire. Occasional managed fires escape control and
disturb large areas. Harris (1966) demonstrated that fire had
little positive effect on regeneration in southeast Alaska, but
Hooven and Black (1978) recommended burning in coastal
Oregon to control pests and suppress competing vegetation.
However, intense burning may destroy mycorrhizal fungi
necessary for effective conifer regeneration (Li and others
1986), and in some cases fire can increase brush competition
by stimulating germination of chaparral seeds.
Burning is increasingly used to reduce fire hazard in
uncut forests. Armour and others (1984) found that understory
community composition in a ponderosa pine forest in northern
Idaho reflected the amount of duff consumed by prescribed
fires, and that pre-burn vegetation communities strongly
influenced the nature of post-burn vegetation. Decreased
fire frequency encourages understory thickets of shade-tolerant
species and accumulation of ground fuels (Parsons and
DeBenedetti 1979). These changes eventually alter species
composition in the forest canopy.
The persistence of a vegetation change depends on an activity’s
duration and on the period required to reestablish pre-disturbance
communities of the original age distribution. Because most
logged sites are not intended to revert to old-growth forest,
changes in vegetation character are effectively permanent. If
species are intentionally removed or added, as by brush control
or establishment of plantations, comprehensive disturbance may
be required to reestablish original communities. In some areas,
intensive removal of a particular species has depleted the reservoir
for recolonization. High-quality cedar is becoming rare in parts of
Washington and is not usually replanted, and Mazurski (1986)
attributed wholesale changes in forest communities in Poland to
five centuries of selective logging.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Timber Management and Soils
Timber management affects soils primarily by compaction,
removal, accumulation, and disruption. Felling and yarding
locally compact soils and expose mineral horizons, with
impact severity dependent on the methods used. Most stud-
ies of soil impacts measure surface disruption or describe
soil properties before and after yarding, and compaction is
usually identified as the most severe physical impact. Extent
of compaction has been measured as a function of location
in the soil profile (Gent and Ballard 1985), soil texture
(Dickerson 1976), moisture content (Moehring and Rawls
1970), type of equipment (Green and Stuart 1985), and
traffic load (Hatchell and others 1970, Sidle and Drlica
1981). Results show that most compaction occurs during the
first few passes, and that its rate and severity are affected by
soil texture. Soil moisture can hasten compaction for some
soils and delay it for others. Hatchell and others (1970)
suggested using different yarding plans to minimize com-
paction on different soil types. Disruption of soil horizons is
particularly severe where site preparation includes soil scari-
fication to expose mineral horizons, or where brush is cleared
by blading or chaining.
How long soil remains compacted after disturbance is
influenced by its texture and by the processes acting to
restore porosity, including root growth, freezing and thawing,
wetting and drying, and animal activity. Each of these
processes is inhibited by compaction, so rehabilitation may
be slow where disturbance is severe. Wert and Thomas
(1981) and Froehlich and others (1985) found that some
soils remain compacted 32 and 23 years after tractor yarding,
respectively, while soils studied by Dickerson (1976) were
estimated to recover in about 12 years. Because of the lengthy
recovery period, both Froehlich (1979) and Geist and others
(1989) considered additional compaction from subsequent
cuts to be a potential cumulative effect. Beasley and Granillo
(1988) expressed concern that the frequency of entry required
by selection cutting may lead to cumulative compaction
because soils cannot recover between entries. Some studies
have measured the effectiveness of artificial means of reducing
compaction (e.g., Gent and Ballard 1985, Voorhees 1983).
Burning can destroy organic-rich horizons and alter soil
structure. Organic loss rates have been measured as functions
of soil moisture, season, and burn intensity (Kauffman and
Martin 1989), and Beaton (1959) described the effects of
fire on soil porosity. Burning makes some soils water-repellent,
which increases runoff and decreases soil-moisture recharge
(e.g., Dyrness 1976, McNabb and others 1989), but burning
enhances infiltration in other soils (Scott 1956). Why soils
become hydrophobic has been studied (e.g., Debano 1981,
Savage 1974), but which soils will be affected cannot yet be
consistently predicted. Durgin (1985) demonstrated that
heating and ash leachates may have opposite effects on soil
dispersivity and erodibility, and that the effects are influenced
by a soil’s stage of development.
Removal of a tree canopy alters microclimates at the
ground surface, and Yin and others (1989) demonstrated
41
severe decreases in decomposition rates where oak forest
floors were exposed by logging in Wisconsin.
Timber Management and Topography
Tractor skid trails and bulldozed fire breaks create depres-
sions that can channel water or intercept shallow subsurface
flow. Where logs are cable-yarded, trails are engraved as
rays radiating upslope or downslope from landings, and
individual trails can generate erosive flows. Cable- or trac-
tor-yarding across low-order and ephemeral channels breaks
down channel banks, and construction of landings usually
requires excavation and emplacement of fill.
Although timber management rarely requires intentional
hydrological changes, minor drainage disruptions can occur
where construction or skid trails divert flow from one drainage
to another. Drainage disruption is avoided when cut units are
carefully laid out, but unmaintained drainage structures may
become plugged and divert flow after a unit is closed.
Topographic effects of timber management were more
severe in the past. During the late 1800s and early 1900s
splash dams were commonly used to create floods to wash
logs downstream (e.g., Wendler and Deschamps 1955), and
channels were cleared to prevent log jams. These effects are
still evident in many west coast streams and must be considered
in CWE analyses, even though the practices were discontinued
long ago. Changes in timber management guidelines during
the 1970s further reduced the extent of topographic disruption
during subsequent logging.
Timber Management and Chemicals
Operation of machinery and chainsaws can lead to spillage of
oil products, and residues remain until washed away or al-
tered by chemical and biological processes. Other chemicals
are added by fertilizers, herbicides, and pesticides. Bengston
(1981) evaluated fertilizer mobility and discussed the cost of
nutrient replacement using several application methods. Most
applied chemicals are quickly metabolized or weathered, and
care is taken to avoid introducing them into streams. How-
ever, even the most careful aerial application is subject to
wind drift. Markin (1982) measured insecticide drifting more
than a mile from an application site and demonstrated that
less than half the applied chemical accumulated on the in-
tended plot. Use of the most harmful pesticides was discon-
tinued in the 1970s, but the effects of earlier practices may
linger because residues of some compounds persist for years.
Use of sewage sludge as a fertilizer and of mill liquor as a
dust suppressant may introduce other compounds that
accumulate in the environment. Brockway (1983) documented
continued elevation of heavy metal concentrations in a pine
forest in Michigan 14 months after sludge was applied, and
Harris and Urie (1983) measured accumulation of nutrients
during 5 years of sludge application in a hardwood forest in
Michigan. Van Cleve and Moore (1978) demonstrated
cumulative changes in soil chemistry and organic matter content
over a 6-year period of fertilizer application in central Alaska.
42
Logging removes nutrients in wood, and in some ecosystems
this represents a high proportion of existing nutrients. Recent
studies documented the effects of tree removal on nutrient
capital (Gholz and others 1985, Mann and others 1988), and
Hornbeck and others (1986) calculated that 20 to 30 years
would be required to reestablish the nutrient balance in a
Vermont forest after logging. Nitrogen and phosphorus are
usually the most limiting nutrients for forest growth. Herbicides
increase nitrogen availability by increasing the volume of
decomposing vegetation (Vitousek and Matson 1985), but
windrowing of slash displaces nutrients to sites inaccessible
for regrowth (Morris and others 1983). Page-Dumroese and
others (1987) explored the effects of mechanical site
preparation in northern Idaho on the distribution of nutrients
and recommended that organics be mixed with topsoil and
mounded into planting beds at nutrient-poor sites.
Burning releases some nutrients to the atmosphere as
volatiles or smoke particles, and alters others into forms
more readily metabolized by plants but more easily leached
from soils. Knight (1966) used laboratory experiments to
show that a fire decreased total nitrogen in a forest floor but
increased the nitrogen available to plants. Little and Ohmann
(1988) and Feller (1988) showed nitrogen loss from burning
in Pacific Northwest forests to be correlated to the proportion
of the forest floor consumed.
Although most impacts from nutrient depletion are expected
only after multiple cutting cycles, long-term effects of timber
management on nutrients have rarely been studied. Results
will undoubtedly vary between sites: Kraemer and Hermann
(1979) showed recovery of nutrient balance 25 years after
logging and burning at 34 sites in western Oregon and
Washington, but Austin and Baisinger (1955) documented
failure of regeneration at multiply-burned sites in western
Oregon where excessive leaching depleted nutrients. The extent
and persistence of chemical changes depend on the types of
compounds involved, soil and vegetation types, and climatic
factors, and are poorly understood.
Storage and milling of logs also introduce chemicals to
the atmosphere and to aquatic systems. Tannins released by
leaching of bark are toxic to aquatic organisms (Temmink
and others 1989) and are likely to reach high concentrations
around log storage decks and in water used to debark logs.
Indirect Effects of Timber Management
on Environmental Parameters
Most impacts described above also influence other environ-
mental characteristics. For example, vegetation changes al-
ter input of organic matter to soils and affect soil structure,
infiltration capacity, and water-holding capacity. Canopy
vegetation moderates forest-floor microclimates, and its re-
moval can increase frequencies and depths of soil freezing
(Pierce and others 1958) and decrease cold-season perme-
ability. Bare soils also experience higher surface tempera-
tures during summer, which can affect soil chemistry, re-
growth, and soil biota.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Vegetation changes may trigger biological succession. Many
early-stage species are capable of fixing nitrogen. However,
silvicultural practice often excludes successional species in
favor of climax conifers and so may impair a forest’s ability
to restore its nitrogen balance after disturbance. Vegetation
changes alter the chemical environment of a site and thus may
affect soil-forming processes, soil biota, and, eventually, the
nature of the soil. Vigorous regrowth induced by logging can
aid recovery of soil structure and reverse the effects of
compaction. Young growth also attracts burrowing animals
that loosen soil and form holes capable of channeling water.
Removal of mature trees alters a forest’s structure and
exposes adjacent trees to windthrow. Studies have evaluated
many factors controlling susceptibility to windthrow, including
age of the cut (Weidman 1920), cutting pattern (Boe 1965),
moisture (Day 1950), topography, and soil type. Schaetzl
and others (1989) reviewed the causes and effects of treethrow.
Uprooting disrupts, mixes, and transports soils, and decreased
windthrow frequency within clearcuts alters soil disturbance
regimes. Armson and Fessenden (1973) predicted accelerated
podzolization with decreased treethrow on clearcuts, and
decreased windfall reduces the volume of woody debris on
forest floors. Downed wood is an important component of
natural forest and aquatic habitats, and these communities
will change if wood inputs decrease.
Establishment of monoculture plantations can make wide
areas susceptible to disease outbreaks, insect infestations, and
other disturbances, but these effects are poorly understood.
Because limited seed stocks were often used in the past for
establishing plantations throughout a region, genetic diversity
of a species may decrease. Reduced variability within a species
can impair its ability to accommodate future environmental
changes and can decrease population fitness by allowing
expression of deleterious recessive genes (Ledig 1986).
Timber-management activities can alter organic matter
content in soils, compact them, expose mineral horizons,
disrupt soil profiles, and alter distribution of soil pores.
These changes, in turn, alter infiltration rates and
permeabilities and influence habitability to vegetation and
fauna. Loss of organic matter by soil displacement or burning
can greatly reduce water-holding capacities of soils. In a
study of regrowth, Duffy and McClurkin (1974) found that
soil bulk density provided the best predictor of regeneration
failure on logged sites in northern Mississippi, and Wert and
Thomas (1981) documented differences in growth rates
between compacted and less disturbed sites in the Oregon
Coast Range. Taylor and others (1966) demonstrated
decreased root penetration with increased soil compaction,
and compaction also hinders root growth by reducing soil
aeration. Moehring and Rawls (1970) found that soil
compaction also reduced growth rates of established loblolly
pines in Arkansas and noted that the only beetle-damaged
trees on their study plots were near soils gouged by machinery.
Soil disruption may have provided pathways for insects to
attack the trees, or the added stress may have lowered the
trees’ resistance. Pre-commercial thinning has been shown
both to increase (Mitchell and others 1983) and decrease
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
(Harrington and others 1983) the incidence of disease and
insect infestation.
Chemical changes usually affect vegetation. A decrease
in nutrient availability can change community composition
by restricting growth rates or selecting for particular species.
Most studies of fertilizer application are short-term and focus
on growth rates of commercial species, but Dunavin and
Lutrick (1983) examined the effects of sludge application on
understory vegetation for many years after application.
Displacement of nutrients, as by windrowing slash, can
accelerate growth rates near accumulations while reducing
growth at depleted sites (Morris and others 1983). Kimmins
(1977) predicted that short-cycle, whole-tree extraction for
fiber use will result in excessive nutrient depletion and outlined
the types of data needed to evaluate potential effects. Long-
term effects of herbicides on community composition are
poorly understood, and McGee and Levy (1988) described
the types of information needed to understand them.
Other impacts related to timber management result from
the influence of logging and associated activities on other
land uses. Logging requires a road network, which increases
recreational use and influences wildfire incidence, and timber
production supports milling and manufacturing industries
and their attendant water developments, population centers,
and potential for degraded air and water quality.
Timber management also influences regional and global-
scale changes that can contribute to CWEs. Burning affects
local climate and contributes to global warming by releasing
carbon dioxide and particulates into the atmosphere. Logging
increases rates of biomass turnover in a forest, and most of
the displaced biomass decays as slash, paper products, and
sawdust; CO2 is released by decay. Harmon and others (1990)
modeled carbon cycling in forests and found that repeated
logging and regrowth in Washington and Oregon has
contributed an annual average of 10 to 20 million tons of
atmospheric carbon for the past 100 years, and thus may be a
significant contributor to global warming. Many parts of the
United States are undergoing net reforestation, however, and
this will counteract some excess CO2 production as long as
standing biomass continues to increase (Sedjo 1990).
Grazing
Many forest lands are grazed, and grasslands often interfinger
with forested tracts, so the effects of grazing must be consid-
ered both in forests and on grasslands. Kauffman and Krueger
(1984) reviewed the effects of grazing on riparian vegeta-
tion, soils, and topography.
Grazing and Vegetation
To understand changes caused by grazing, natural patterns
of rangeland disturbance must first be understood. Charac-
teristic fire frequencies have been measured for some grass-
43
lands using scarred trees (Arno and Gruell 1983) or historic
records (Higgins 1984), but data are rare because fire scars
are poorly preserved. Drought is also an important distur-
bance. Murphy (1970) developed a method to predict Coast
Range grass yields from fall precipitation, and Gibbens and
Beck (1988) described temporal patterns of rangeland re-
sponse to drought and demonstrated differences between
annual and perennial species. Descriptions of drought ef-
fects are common (e.g., Ellison and Woolfolk 1937).
Most rangeland impacts are caused by grazing, and most
grazing alters vegetation. Each species of herbivore prefers
certain plants, and these are depleted on pastures grazed by
that species. Thus, grass density tends to decrease on cattle
allotments in Utah and to increase on sheep allotments (Shupe
and Brotherson 1985). Ruyle and Bowns (1985) suggested
that complementary grazers be pastured together to balance
forage use and preserve the original plant community.
Grazing preferences and trampling can prevent regeneration
of some species, and the sparsity of young oaks in California
savannas is thought by some to reflect grazing pressure (e.g.,
Rossi 1980) or changes in grass vegetation caused by grazing
(Danielsen and Halvorson 1991). California Coast Range
grasslands changed from predominantly perennial bunchgrasses
to European annuals with the introduction of cattle (Burcham
1982), and bunchgrasses now are dominant only at sites
protected from grazing. Annual grasses replaced annual forbs
in Central Valley prairies (Wester 1981), and shrub and tree
densities have increased in interior rangelands (Branson 1985,
Herbel 1985). The relative importance of grazing, logging,
changes in fire frequency, and loss of endemic herbivores in
causing these changes is not known, but the overall impact is
a cumulative effect of the combination of uses. Grazing gives
a competitive advantage to star thistle, artichoke, tarweed,
and other plants associated with disturbed soils.
Although grazing allotments are granted on the basis of
average stocking rates, use of the range is not uniform.
Roath and Krueger (1982) reported that 81 percent of forage
use was sustained by a riparian corridor that occupied only
1.9 percent of their Oregon study area. Senft and others
(1985), Mitchell and Rodgers (1985), and Marlow and
Pogacnik (1986) showed seasonal differences in the types of
plants and sites used by cattle, with increased use of browse
plants and riparian zones as the dry season progressed. Gillen
and others (1985) explored the influences of different grazing
strategies on riparian habitat use.
Heavy grazing decreases ground-cover density both by
consumption and by trampling, and Abdel-Magid and others
(1987b) measured these effects in a laboratory. Olson and
others (1985) examined the interaction between grazing
intensity and precipitation and showed that their effects on
ground cover density varied with the type of plant. Vegetation
can be completely absent along cattle tracks, in holding
pens, and near water sources, where use intensities are high.
Riparian vegetation is particularly heavily used and is often
noticeably sparser than in exclosures (Kauffman and others
1983b, Platts and others 1983), and Kauffman and others
(1983b) showed that grazing retarded succession on channel
44
bars in northeast Oregon. Schulz and Leininger (1990)
compared vegetation inside and outside a 30-year-old riparian
exclosure in Colorado to demonstrate decreased total cover,
decreased peak standing crop, increased exposure of bare soil,
and altered community composition due to grazing.
Damage to cryptogam communities is less visible. These
alliances of algae, lichen and moss reduce erosion and promote
infiltration by coating bare soil surfaces (Loope and Gifford
1972). Trampling destroys fragile cryptogam crusts, and
Anderson and others (1982) measured a 50 percent decrease in
cryptogam cover at some grazed sites in Utah. Little is known
of recovery rates after grazing, but Anderson and others (1982)
documented continued recovery after 15 years of exclosure.
Even less evident is grazing-related damage to the mycorrhizal
fungi that aid plant growth and stabilize erosive soils. These
fungi rely on certain vascular plants for nutrients, and grazing
can destroy mycorrhizal communities by suppressing their host
plants (Bethlenfalvay and Dakessian 1984).
Grazing strategy influences the types of changes caused
by grazing, although Van Poollen and Lacey (1978) compiled
published research from many sites to demonstrate that
stocking rate is a more important influence. Managers can
select grazing intensities, rotation cycles, and seasonality of
use, and much research has been devoted to quantifying the
effects of these variables on forage production, animal
nutrition, and species composition (e.g., Heady 1961, Wood
and Blackburn 1984). Platou and Tueller (1985) argued that
natural grazing patterns in plant communities should indicate
the appropriate grazing strategies for those communities,
with Great Basin communities responding best to rest-rotation
plans while environments of the Great Plains are better
suited to high-intensity, short-duration grazing.
Range-improvement plans are usually designed to alter
vegetation to more palatable species, and may include weed-
eradication by herbicide or physical removal, fertilization,
mechanical brush removal, plowing, tree thinning, and
prescribed burning. The benefits and long-term effects of
some widely used practices remain poorly documented. Rippel
and others (1983) found that tree and shrub densities 20
years after brush removal in a pinyon-juniper community in
New Mexico were approximately the same as before, and
that grass and forb densities were lower. Johansen and others
(1982) showed that burning for range improvement in Utah
altered perennial age structures and decreased cryptogam
covers, though apparently without major changes to the
community structure. Towne and Owensby (1984) explored
the effects of seasonality of burning on vegetation
communities and used this information to design a burning
plan for a Kansas prairie.
Recovery rates for vegetation communities must be
determined if the cumulative impacts of grazing are to be
predicted. Gambel oak brushlands studied by Austin and
others (1986) in Utah had recovered from grazing impacts
by 30 years after abandonment. Short-term studies indicated
significant improvements in riparian vegetation at an eastern
Oregon site within 3 years (Kauffman and others 1983b) and
partial recovery of woody vegetation in 10 years (Rickard
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
and Cushing 1982). Dry grasslands in Utah accomplished
only 40 percent of their recovery in ground cover density
after 10 years of exclosure (Kleiner 1983). In other cases,
recovery may not be possible. Brushlands exclosed from
grazing for 13 years in central Utah showed no tendency to
reestablish their original sagebrush-grassland community
(West and others 1984), suggesting that this grazing-induced
vegetation change is stable under present conditions.
Collins (1987) explored the combined effects of grazing
and fire on a prairie community and found that each
disturbance influenced the response to the other. This type
of synergistic effect is likely for most combinations of
ecosystem disturbance, but has rarely been studied.
Grazing and Soils
Heavy trampling can compact subsurface layers while loos-
ening the soil surface, and many studies have examined the
influence of stocking rates, seasonality, soil moisture and
soil texture on compaction (e.g., Abdel-Magid and others
1987a; Warren and others 1986a, 1986b, 1986c). Compac-
tion generally increases with stocking rate and is most pro-
nounced on moist, fine-textured soils. Cattle trails are par-
ticularly heavily impacted, and Walker and Heitschmidt (1986)
described variation in trail density as a function of grazing
strategy and proximity to a water source. Abdel-Magid and
others (1987b) measured average daily travel distance for
cattle and found it to depend on grazing strategy. Persistence
of compaction has rarely been measured. Stephenson and
Veigel (1987) demonstrated a 92 percent recovery of bulk
density in a fine loam after 2 years of exclosure in southern
Idaho, and Bohn and Buckhouse (1985) showed that rest-
rotation grazing allowed partial recovery of soil conditions
on a sandy loam soil in northeastern Oregon while deferred
rotation (grazing alternate years) did not. Trampling of wet
soils was found to be particularly detrimental.
Vegetation changes also affect soil properties. Plowing to
suppress brush increases soil surface roughness and may
improve infiltration, while burning can cause water repellency
(Brown and others 1985). Over a long period, vegetation
change by grazing or range improvement can modify soil-
forming processes and change chemical and physical
properties of soils.
Grazing and Topography
Grazing animals often follow hillslope contours as they
roam, and repeated grazing creates parallel trails on hillsides
that lend them a staircase-like profile. On a finer scale,
trampling on wet soils engraves footprints in the mud and
increases soil-surface roughness, while trampling on dry
surfaces breaks down irregularities. Animals avoid stepping
on grass tussocks, and channeling of traffic around tussocks
can increase elevation contrasts around the plants (Balph
and Malecheck 1985). These changes in surface roughness
affect rates and erosivity of overland flow.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Topography is often altered by construction of stock-
watering ponds. These small dams decrease downstream
flows early and late in the rainy season and modify peaks
slightly during most storms. Streams are occasionally diverted
for water sources. Trampling by grazing animals strongly
influences channel morphology, and channels in grazed areas
tend to be abnormally wide and shallow (Platts and others
1983) and have a lower proportion of overhanging banks
(Kauffman and others 1983a).
Other Effects of Grazing
Use of fertilizers or herbicides for range improvement adds
chemicals to the environment, but range management of this
intensity is not common in forest lands. Grazing redistrib-
utes nutrients as manure and makes some nutrients more
accessible for plant use. Manure often increases nutrient
inputs to streams and can contribute to salinization.
Fecal coliform counts in streams usually increase with
grazing intensity, although Buckhouse and Gifford (1976)
showed no significant change in fecal coliforms with moderate
grazing intensities in Utah. Coliforms are important as indicators
of accompanying pathogens (Van Donsal and Geldreich 1971).
Coliform counts are strongly influenced by the distribution of
animal use in relation to channels, and Tiedemann and others
(1987) showed that coliform concentrations varied with grazing
strategy in eastern Oregon. Tiedemann and others (1987) also
found that coliforms could overwinter at a site in the absence
of herbivores, and Jawson and others (1982) measured elevated
bacterial levels at a site in Washington that persisted until 2
years after grazing ended. Stephenson and Rychert (1982)
measured high coliform concentrations in stream sediments in
a grazed area of southern Idaho.
Indirect Effects of Grazing on
Environmental Parameters
Grazing influences many environmental parameters indi-
rectly. Altered vegetation affects soil porosity, texture, and
organic content by changing the rate of litter accumulation
and the distribution and longevity of roots, and suppression
of cryptogam and mycorrhizal communities changes soil
texture and surface roughness and retards nitrogen fixation.
Vegetation change eventually alters soil-forming processes
and may change the type of soil present. Vegetation changes
can provoke further successional change, which may com-
pletely alter plant communities. This is the objective of
vegetation management, and Sturges (1983) showed that
herbage production maintained a 100 percent increase 10
years after sagebrush had been removed from a site in Wyo-
ming. Changes in bulk density, pore distribution, and or-
ganic content alter infiltration rate, soil strength, and water-
holding capacity, and so modify a soil’s ability to support
particular plant species. Compaction inhibits growth in many
species, and Rickard (1985) described replacement of native
plant communities by alien annuals at trampled sites in
eastern Washington. Trampling of streambanks often causes
45
channel widening and destroys riparian vegetation.
Most chemicals applied to rangelands are intended to
alter vegetation. Fertilizers increase biomass and cover density,
while herbicides suppress particular species, and extensive
research is directed at enabling prediction of these effects
(e.g., Wight and Godfrey 1985). Grazing animals preferentially
occupy riparian areas, and their manure can increase stream
nutrient loads and cause algal blooms and eutrophication.
Other land uses associated with grazing also influence
environmental parameters. In particular, new grazing access
roads open remote areas to recreational use; impoundments
are constructed for watering stock; and channels may be
diverted or wells drilled as water sources.
Interactions Between Grazing and
Timber Management
Grazing and timber management are frequently associated,
and interactions between these uses are likely causes for
CWEs. Multiple-use mandates often provide for coexistence
of these uses at a site, and grazing is sometimes used as a
timber-management tool. In other cases, grazing and log-
ging are carried out in different parts of a watershed and both
can contribute to the same downstream impact.
Foresters occasionally prescribe grazing to decrease fuel
loading and reduce fire hazard. Removal of fine fuels by
grazing may decrease the effectiveness of prescribed burns
at some sites, however, and thereby increase long-term fire
danger (Zimmerman and Neuenschwander 1983, 1984).
Grazing is also prescribed to suppress brush and reduce
competition for young trees. Richmond (1983) showed
increased radial growth rates of 7 to 14 percent for Douglas-
fir at a site grazed by sheep in the Oregon Coast Range.
However, grazing must be managed carefully if impacts on
timber production are to be minimized. Most studies indicate
that moderate grazing does not hinder conifer regeneration
(e.g., Kosco and Bartolome 1983), but Eissenstat and others
(1982) found that 19 percent of seedlings in a cattle-grazed
Douglas-fir plantation in northern Idaho had been trampled
and suggested that grazing should be excluded from newly
planted sites. Studies of seasonal diets suggest that use patterns
may be adjusted to remove particular plant species (e.g.,
Fitzgerald and Bailey 1984, Leininger and Sharrow 1987).
Kingery (1983) discussed interactions between timber
management and grazing and indicated the need for data
showing which strategies are effective in which settings.
Mining
Underground mines now represent only a minor part of the
minerals industry in the United States, and most income and
environmental change result from oil and gas extraction,
open-pit mines, strip mines, and gravel pits. Mining of sand,
gravel, and rock for construction materials is the largest non-
46
fuel component of the U.S. mining industry (Hoffman 1990).
In-stream gold dredging is common as a quasi-recreational
pastime in parts of California.
Mining and Vegetation
Open-pit and strip mines denude large areas of vegetation,
and oil fields require extensive road networks and cleared
pads for access to wells. Gravel mines often remove riparian
vegetation from bars, floodplains, and terraces. Accumulat-
ing deposits, such as bars, are usually stripped annually,
while terraces and other relic deposits are progressively
cleared. In-stream gold dredging disrupts stream-bed plant
communities and can increase downstream bacterial counts
by releasing bacteria associated with bottom sediments
(Grimes 1975).
Mine dumps are often difficult to revegetate, and many
studies test methods to improve growth on rock waste and
refilled mine pits. Holechek and others (1982) tested the use
of fertilizers and warned against planting aggressive exotic
plants. Stark and Redente (1985) correlated revegetation
success on simulated oil-shale waste with various site and
soil conditions in northwest Colorado, and McGinnies (1987)
measured the effects of mulching on coal mine tailings in
Colorado. Power and others (1981) tested different depths of
topsoil and subsoil to determine optimal thickness for
promoting plant growth on reclaimed mine land in North
Dakota. Exposure of mineral soil retards colonization of
higher plants by suppressing soil mycorrhizae. Lindsey and
others (1977) noted the importance of mycorrhizae to growth
and survival of native species, and attributed the effect to
increased nutrient uptake. Reeves and others (1979) found
that 99 percent of plants in undisturbed Great Basin
communities were mycorrhizal, while only 1 percent of
those at disturbed sites were. Doerr and others (1984) noted
that severity of soil disturbance influenced the type of
regrowth: highly disturbed substrates no longer had the
mycorrhizae needed to support native perennials and were
instead colonized by annuals.
Other changes to vegetation are indirect. Over- or
undercompaction of tailings piles inhibits revegetation by
decreasing infiltration, suppressing root growth, or creating
overdrained substrates. Topographic changes can retard
revegetation if slopes are steep enough to permit ravelling of
surface sediments. Unvegetated tailings piles can be abnormally
erodible because surfaces are not protected by vegetation and
because developing soils are low in organic content and are
less structured than those influenced by vegetation.
Toxic chemicals introduced into streams by mining can
suppress floodplain vegetation and promote bank erosion
(Lewin and others 1977), and atmospheric mill emissions
contributed to devegetation of 125 km2 of forestland at
Sudbury, Ontario, and damaged plants on another 500 km2
(Pearce 1976). Amiro and Courtin (1981) described the
vestigial plant communities remaining there. Even where
chemicals do not destroy vegetation, they can be assimilated
by plants and passed into the food chain. Neuman and
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Munshower (1984), for example, showed that high
molybdenum contents in coal mine soils led to copper
deficiencies in cattle.
Mining and Soils
Tailings piles range from overdrained to overcompacted;
rarely are permeabilities and bulk densities characteristic of
natural sites. Reconstructed topsoil density on reclaimed
mine lands must be similar to surrounding soils to avoid
runoff and landslide problems. Machinery used in pit and
strip mines compacts trafficked surfaces, and chemicals in
some spoils piles alter their hydrologic character. Gilley and
others (1977) described coal mine dumps in North Dakota
where exposure of dispersive sodic materials rapidly sealed
the soil surface and prevented infiltration.
Surface mining removes or disrupts soils. Gravel mines,
for example, prevent bar stabilization and floodplain soil
formation by repeatedly removing point-bar deposits and
disrupting incipient soils. Gold dredging of the past century
covered many floodplains with porous gravel deposits, leaving
broad areas denuded along rivers in California’s Central Valley.
Mining and Topography
Historic mining left an imprint of holes, mounds, and rhyth-
mically ridged floodplains over much of the western United
States, and most of these topographic changes can reroute
low-order channels. Hydraulic mining in California removed
hills, filled valleys, and caused impacts that provoked the first
formal CWE analysis and are still felt today. Mine and quarry
faces and tailings dumps often have been left at steep angles
that promote rockfall, landsliding, and surface erosion.
In-channel dredging and stream-side mines alter channel
morphology, and gravel mines on floodplains excavate pits
that are later reoccupied by channels. These effects were
reviewed by Sandecki (1989), who noted their cumulative
nature. Bull and Scott (1974) described a flood on Tujunga
Wash during which the channel was rerouted by overtopping
of gravel pits and destroyed a long section of highway.
During lower flows, pits can prevent infiltration through the
channel bed by ponding water and allowing fine sediments
to settle (Bull and Scott 1974). Local stream-bed excavation
can trigger upstream incision by knickpoint retreat, and
Harvey and Schumm (1987) and Collins and Dunne (1989)
documented cases where continued extraction of bed materials
resulted in channel incision. Floodplain vegetation helps
stabilize stream banks, and its removal by gravel mining can
provoke bank erosion and channel widening. Despite the
importance of gravel mining in California, little is known of
its impacts on channel morphology.
In-channel dredging generates high turbidities from
disturbance of streambed sediments. Reynolds and others
(1989) measured five-fold increases in suspended sediment
concentrations downstream of gold dredges in Alaska. Fine
sediments are usually redeposited downstream, and Bjerklie
and LaPerriere (1985) described the resulting decrease in
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
bed infiltration and lowering of stream-side water tables.
Hydraulically unstable mounds of gravel left by dredging
are easily rearranged during high flows, and the increased
gravel transport can destabilize banks, widen channels, and
disrupt floodplains.
Mining and Chemicals
Mining alters the chemical environment by releasing wastes
and by exposing weatherable material. Erdman and others
(1978) measured molybdenum enrichment in vegetation grow-
ing on coal spoils piles in the Great Plains and noted that
levels were high enough to poison livestock. Tailings leachates
may contain toxic pollutants, and acid mine drainage can
severely lower pH in streams. Collier and others (1964)
found pH levels lethal to fish downstream of a Kentucky
coal mine. Streams near cobalt-arsenide mines in Idaho are
heavily contaminated by arsenic from the ore, and have also
been acidified by weathering of newly exposed sulfides
(Mok and Wai 1989). Drainage from a uranium mine intro-
duced cadmium, strontium, and uranium into streams (Nichols
and Scholz 1989), and placer mining for gold can mobilize
heavy metals (LaPerriere and others 1985). Activities asso-
ciated with mining also have chemical effects: processing
mills often release atmospheric contaminants, and oil carri-
ers and pipelines are subject to leaks.
Interactions Between Mining and
Timber Management
The high demand for timbers to shore up early underground
mines deforested large areas in the western United States,
and some of these sites have not fully recovered. Smelter
emissions have also taken large areas out of forest produc-
tion (Amiro and Courtin 1981, Pearce 1976).
Sand and gravel mining is common on California forest
lands, and is also usually present downstream of forested
watersheds. Gravel mining has even been suggested as a
mitigating measure for excessive sediment input from earlier
logging. In-stream gold dredging is common along forested
streams of northwest California and the Sierra Nevada.
Agriculture
Agriculture often occupies bottomlands in forested water-
sheds. Agricultural use includes soil preparation, planting,
irrigation, fertilization, pest control, weed control, and har-
vesting of field crops, row crops, and orchards.
Agriculture replaces natural vegetation with cultivated
species and periodically removes the crop. Ground cover
densities may be higher than natural while field crops are
growing, but are usually low after tillage, and Wischmeier
and Smith (1978) compiled measurements of seasonal cover
densities for many crops. Introduced crop species occasionally
47
thrive in the wild and may threaten other uses. Wild artichoke
decreases range productivity, for example, and teasel, mustard,
and wild radish are considered undesirable weeds.
Tillage usually reduces compaction near the soil surface
(Voorhees 1983), but compaction may increase below the
plow layer (Blake and others 1976). Hill (1990) found lower
bulk densities throughout the soil column in conventionally
tilled plots than in no-till plots on a Maryland silt loam.
Agricultural servicing areas are usually highly compacted
and have correspondingly low infiltration rates. Soil
preparation activities are designed to make soils more
workable and often increase their organic content. Crop
residues and imported organic matter may be plowed back
into the soil, and chemicals may be added to alter soil
structure. If plants are harvested without reincorporating
organic materials, however, soil organic matter may be
depleted. Woods and Schuman (1988) reported lower organic
matter content in cropped soils than in the same soils without
crops at a site in Colorado and showed that much of the
change occurred during the first year of cultivation.
Repeated fertilization and cropping can modify soil
structure, and structure can be improved by planting forage
crops (Perfect and others 1990). Anderson and others (1990)
measured decreased bulk density and increased water
retention, organic matter, and pore size in Missouri fields
manured annually for 100 years. Soil changes may persist
over long periods and affect vegetation communities
regrowing at the site. Kalisz (1986) measured abnormally
coarse soil textures more than 60 years after eroded agricultural
land was abandoned in Kentucky. Plowing of shallow soils
underlain by calcrete can mix the calcified layer into the
topsoil and damage its texture and productivity, but deep
ripping can improve soil drainage by breaking up hardpans.
Repeated plowing around obstructions or along fence
lines can cause topographic discontinuities to grow because
most plows displace soil to one side. Terracing for erosion
control and access modifies profiles of steep slopes, but
even terrace faces are rarely steep enough to be unstable in
areas of mechanized agriculture. On a smaller scale, plowing
roughens the soil surface. Some crops are planted in furrows
or on ridges or mounds, while paddy rice requires ponded
depressions. Surface flow within fields is controlled by
drainage and irrigation ditches, which must be carefully
designed to prevent incision, aggradation, or widening.
Reclamation for agricultural development has severely
modified channel morphology in fertile lowland areas. In the
Willamette Valley, for example, the complex network of
bayous and floodplain channels that once occupied the valley
floor has been reduced by land reclamation to a single
meandering channel (Sedell and Froggatt 1984). Changes in
California’s Central Valley are as extreme. Valley-wide
seasonal marshes have been drained for farmland, and mainstem
channels are dredged and revetted to contain flows. Seasonally
flooded islands of the Sacramento River delta are now leveed
and drained, and progressive drying and oxidation of their
peat substrates has caused cumulative subsidence. Floodplains
48
are often the most desirable sites for farmland, and channels
are commonly engineered to reduce flooding, bank erosion,
and channel migration. Modifications include bank protection,
levees, and dredging. Many California crops are irrigated and
require a complex infrastructure of reservoirs and canals.
Chemicals are widely used in agriculture as fertilizers,
pesticides, and herbicides, and each chemical has a different
mobility and persistence in the environment. In California,
10 percent of tested wells had higher nitrate concentrations
than standards permit, and most contamination came from
agricultural fertilizers and livestock waste (Mackay and Smith
1990). Pesticide contamination was also frequent (Mackay
and Smith 1990). Many contaminants have long been out of
use but persist in groundwater and stored sediment (Gilliom
and Clifton 1990).
Harvesting removes nutrients from fields at rates that vary
with crop type, harvesting method, and the fate of crop residues.
Nutrient loss is reduced where residues are left in fields, but
where residues are burned, some nutrients are volatilized and
lost while others alter to forms both more accessible to plants
and more readily lost by leaching. Burning of residues
introduces particulates into the atmosphere and contributes to
carbon dioxide accumulation, but the practice continues as an
inexpensive control for many agricultural pests.
Irrigation alters soil chemistry by mobilizing and
precipitating soil salts, and salt may also be imported with
irrigation water. Vast tracts have been made unarable by salt
accumulation, and irrigation effluent salinizes streams in many
semiarid areas.
Irrigation developments often move water from one
watershed to another in natural channels or constructed canals.
Water imports increase soil moisture in irrigated fields, permit
higher evapotranspiration rates, and may locally raise water
tables. Areas from which water is diverted experience the
opposite effects.
Logging and agriculture are often closely linked. Woodlots
often abut fields in main-stem river valleys, and water from
most forested land eventually flows through agricultural areas.
Logging has historically accompanied agricultural clearing.
In many parts of the nation, once-tilled lands are now reverting
to forest, but nutrient depletion and soil changes from cropping
influence the regenerating communities. Deforestation for
agricultural development in western Australia has decreased
evapotranspiration and raised valley-bottom water tables,
bringing salts to the surface and preventing further agriculture.
Measures to reduce salinity now include reforestation to
lower the water tables (Bell and others 1990).
Urbanization
Growth of urban and suburban communities is rapid in the
western United States and has affected many watersheds
that historically supported only resource-based activities.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Vacation and retirement homes have long occupied wildland
margins, but now rising costs of urban living have made 1-
to 2-hour commutes economically feasible, and primary homes
are increasingly common at wildland margins. In addition,
regional economic centers have grown to become commute
destinations with their own suburban fringes.
Urbanization includes construction activities as well as
industrial and residential use. Construction rates are episodic
at most sites; some population centers are now stable in size,
while others undergo exponential growth. Wolman and Schick
(1967) provided estimates of construction area per unit
population increase.
Urbanization is accompanied by power development.
Hydroelectric dams and diversions, fossil-fuel plants, and
nuclear reactors provide much of California’s power, and
wind and solar power are increasingly used in some areas.
Power generation may be on the scale of domestic or indus-
trial supply, or may merely provide energy for an internal
combustion engine. Majumdar and others (1987) edited a
volume that examines environmental impacts caused by power
development, and El-Hinnawi (1981) reviewed a variety of
impacts from energy production.
Urbanization and Vegetation
Urban growth entails comprehensive vegetation change.
Construction removes vegetation over much of a project
area, and native species are replaced by introduced plants
that often require irrigation and pesticides. Native shrubs are
often removed to lessen fire danger in chaparral areas. Al-
tered vegetation cover may locally be more continuous than
the original cover, but much of a developed area is covered
by asphalt and structures, and unpaved ground subjected to
traffic is usually left bare.
Where natural vegetation is preserved, it is often in
unnatural contexts. For example, plant communities in isolated
parks and natural areas change character because they no
long sustain natural grazing pressures and disturbance
frequencies, and chaparral stands preserved in the absence
of wildfire accumulate catastrophic levels of fuel. Wilson
and Ferguson (1986) found that fire intensity was the best
predictor of damage to structures in developed chaparral
areas, and implied that fire intensity was a function of fuel
loading. Even in low-density vacation settlements, vegetation
is often manipulated to reproduce a more urban flora. Clark
and Euler (1984) documented this change around a rural
vacation development in Ontario.
Introduced plants may escape from cultivation and alter
community composition in surrounding areas. Ornamental
foxglove, for example, has become a common component of
coastal forests, and introduced pampas grass inhibits conifer
regeneration on clearcuts in some areas. Increased nutrient
loads in urban streams and lakes can trigger rapid growth of
algae (Welch and others 1989), which depletes dissolved
oxygen and decreases penetration by light.
Changes in soil structure can alter vegetation communities,
and many imported species have a competitive advantage on
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
disturbed sites. Where slopes are steepened by excavation,
continued ravelling can prevent revegetation. Riparian
vegetation can be modified by channelization near urban areas
and by reduced or fluctuating flows below hydroelectric projects.
Urban smog is associated with damage to natural
vegetation. Omart and Williams (1979) measured decreased
growth rates in smog-damaged trees in southern California.
Treshow and Stewart (1973) exposed understory plants to
various concentrations of atmospheric pollutants for 2 hours,
and of the 70 species tested, only five emerged undamaged.
Urbanization and Chemicals
Urban and suburban land use alters the chemical environ-
ment in many ways. Oil products collect on road surfaces and
are washed into streams with runoff. Domestic and industrial
operations make heavy use of environmentally toxic sub-
stances such as polychlorinated biphenyls (PCBs), pesticides,
herbicides, and solvents, and many of these enter sewer
systems. Past disposal practices for these hazardous materi-
als are now haunting communities with area closures, envi-
ronmental illness, and costly cleanups. Between 5 and 20
million gallons of oil have been spilled on U.S. waters each
year during the 1980s (Reid and Trexler 1991), underground
fuel storage tanks are increasingly found to be leaking into
groundwater, and toxic spills have become common.
Sewage disposal often increases phosphorus and nitrogen
loads in waterways. Welch and others (1989) documented
phosphorus enrichment downstream of urban areas on the
Spokane River and related phosphorus levels to algal biomass.
Gold and others (1990) showed nitrate inputs in areas served
by septic tanks to be as high as those in agricultural areas, and
Osborne and Wiley (1988) correlated phosphorus and nitrogen
content in streams with the percentage of basin area urbanized.
The major impacts from fossil-fuel-burning power sources
are caused by the wastes they emit: smoke particles, toxic
compounds, and chemicals that react with other atmospheric
constituents. Some components of smog increase the acidity
of precipitation, and thus of soils and surface waters. Others
weaken the atmospheric ozone layer and permit more
ultraviolet radiation to reach the earth’s surface. Carbon
dioxide released by burning may increase atmospheric
insulation, raising global temperatures and altering weather
patterns. Hansen and others (1987) predicted a 3° average
temperature rise with the expected doubling of atmospheric
carbon dioxide during the next century. Global warming
may eventually destabilize polar ice and raise sea level
enough to flood coastal communities.
Other Effects of Urbanization
Construction activities compact soils and remove organic
horizons. In residential areas, natural soils may be replaced
by more workable loams or altered by addition of soil
amendments. Developed sites include impermeable surfaces
such as roofs, roads, and parking lots, and much of the land
surface is impermeable in heavily urbanized areas. Stankowski
49
(1972) presented a method for calculating percent
impermeable area on the basis of population density. Soil
compaction on bare surfaces may decrease permeabilities in
rural developments. Removal of natural vegetation at building
sites can alter soil strength and contribute to landsliding.
Urban hills are leveled for building sites, valleys filled
for road access, and slopes cut back to provide level ground.
Channel networks are often completely restructured, and
streams are replaced by sewer systems, conduits, and concrete
channels to hasten removal of peak flows. Graf (1977)
described topographic changes to stream networks in a
developing area near Iowa City and demonstrated a 50 percent
increase in drainage density due to artificial channels. Urban
river banks are usually rip-rapped or reinforced to prevent
channel migration and leveed to prevent flooding.
Walter and others (1989) showed virus concentrations in
two German rivers to be proportional to sewage inputs, and
Cowan and others (1989) found increased levels of fecal
coliforms in streams draining residential and industrial areas
in Newfoundland. Other organisms also accompany
urbanization: introduced rats, mongooses, cats, starlings,
and pigeons have all affected native biota, and urban fauna
have influenced world history as vectors for human disease.
Plant diseases and pests are also imported and spread by
introduced species. Replacement of transpiring vegetation
by impervious surfaces alters local climates, as does addition
of atmospheric particulates.
Hydroelectric dams usually release flows during times of
high power demand, and fluctuating outflows can destabilize
channels by increasing pore pressure fluctuations in
downstream channel banks. Where penstocks transport
reservoir outflow to increase power-generating head, flows
are reduced to part of the downstream channel.
Nuclear power plants create radioactive waste and excess
heat that must be disposed of. There are occasional reports
of radioactive groundwater contamination at the Hanford
Nuclear Reservation in Washington State, and other storage
sites for low-level radioactive wastes are likely to be subject
to similar leakage. Large volumes of water are used to cool
reactors, and the thermal plume introduced to the Columbia
River from Hanford is visible on infrared aerial photographs.
Recent incidents at Three Mile Island and Chernobyl have
renewed public concern over the susceptibility of nuclear
power plants to catastrophic failure and radioactive releases.
Interactions Between Urbanization
and Timber Management
The timber industry employs many people and requires infra-
structure support from many others, and logging-supported
communities have grown with the timber industry. Recre-
ation-based communities are also common in forested water-
sheds. Urbanization increases timber demands because of its
requirements for building material and paper products. For-
est lands surrounding developing population centers are pro-
gressively cleared as the communities grow, and concerns of
residents can limit timber management on nearby lands.
50
Flood Control and Navigation
Flood control programs include construction of impound-
ments, levees, and overflow channels to contain flows, and
dredging and straightening of channels to increase flow
capacity. Navigational modifications to improve or maintain
water courses are similar, and include dredging, meander
cutoffs, bank protection, construction of locks, and removal
of logs and obstructions.
Conveyance and navigability of natural channels are often
increased by removing drift logs, and channels modified for
flood control may be cleared of obstructing riparian vegetation.
Riparian vegetation is also destroyed when channel
morphology or banks are altered, and regrowth is prevented
by bank protection works designed to maintain the channel’s
new form. Vegetation can be altered over wide areas where
flooding is controlled by temporary filling of overflow
channels or leveed lowland basins.
Dredging disrupts in-channel vegetation (Pearson and
Jones 1975), and Grimes (1975) found that navigational
dredging increased concentrations of fecal coliforms by
resuspending the sediments with which they are associated.
Dredge spoils are commonly placed on floodplains or are
used to augment levees. Spoils piles are often difficult to
revegetate and must be carefully placed to avoid reintroducing
sediments to the channel. Other effects of dredging are
similar to those caused by in-stream mining.
Levee construction and channel engineering usually involve
replacement of near-channel soils with riprap, grout, or
revetments to protect banks. Soils in areas protected from
flooding are no longer replenished by sediment and organic
material carried in by flood waters, and channelization in
estuaries and deltas can lead to erosion of unreplenished
deposits.
Commercially trafficked waterways like the Mississippi
have undergone major morphological alteration: the planform
is straightened, the thalweg dredged, and the long-profile
broken by locks. Many of the changes described by Sedell
and Froggatt (1984) on the Willamette River in Oregon were
carried out to improve navigability. Canals are often dug to
avoid obstructions or connect waterways, and these alter
network configurations and allow passage of biota from one
system to another. Some species are readily transported on
the hulls of ships.
Flood control projects almost always involve direct
topographical changes: channels are moved and reshaped, bed
topography is altered, and levees are constructed. Each of these
affects the velocity and depth of flow through the altered reach
and the rate at which water is delivered to reaches downstream.
Desiccation of reclaimed floodplain soils can cause land to
subside and may induce even more catastrophic flooding should
levees fail and newly settled land again be immersed. Where
flood diversions return flow to a different watershed, source
channels experience reduced runoff and peak flows, while
target channels must adjust to increased flows.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Boat wakes increase turbidity by eroding banks and
disrupting bottom sediment, and their turbulence can keep
sediment suspended. Garrad and Hey (1988) measured diurnal
turbidity fluctuations associated with use levels in trafficked
waterways. Increased turbidity decreases light penetration,
inhibiting growth of macrophytes in the waterway and thus
further promoting erosion (Garrad and Hay 1988).
Chemicals and nutrients can be introduced to channel
systems if flood control measures include overflow drains
on urban sewer systems, or if overflow basins are seasonally
cropped with the aid of herbicides or pesticides. Urban flood
drainage is repeatedly implicated for catastrophic fish kills
on eastern rivers. If flood waters are stored in shallow basins,
primary productivity in the temporary ponds can be high
enough to increase the nutrient content of returning water.
Boat traffic can introduce pollutants to waterways. Two-
cycle outboard engines release up to half of their fuel unburned
(Liddle and Scorgie 1980), and anti-fouling paints contribute
to toxic conditions around marinas and harbors. Oil slicks on
lake surfaces inhibit gas exchange to the lake waters (Barton
1969). Turbulence from wakes can increase phosphorus
activity by resuspending bottom sediments (Yousef and others
1980), and may contribute to eutrophication.
Recreation and Fishing
Recreational uses of wildlands range from hiking, fishing,
skiing, and rock-climbing to water skiing, trail biking, and
driving off-road vehicles (ORVs). Developments such as
resorts, ski areas, and campgrounds are designed to encour-
age recreational use, while new roads into remote areas
promote recreation as an unintended by-product of develop-
ment. In other cases, lack of development is itself the recre-
ational attraction, and wilderness areas and wild and scenic
rivers become foci for those seeking experiences in nature.
Boden and Ovington (1973) noted that areally averaged
visitor use in undeveloped areas misrepresents actual use
intensities because use is concentrated in a small proportion
of the area. Wall and Wright (1977) reviewed the impacts of
recreation on vegetation, soil, water quality, and wildlife.
Vegetation is altered both by development of focal sites
and by repeated use of dispersed sites. Hikers trample some
meadows to the point that they are no longer viable (Willard
and Marr 1971), pack-stock overgraze meadows and riparian
zones, and ORVs create bare tracks (Davidson and Fox
1974). Some species and vegetation communities are more
susceptible to trampling than others. Cole (1978) found that
vegetation in dense forests in northeast Oregon is more
sensitive to change than in meadows or open forest, and Bell
and Bliss (1973) showed impacts on alpine stone-stripe
vegetation to be more severe than on snow-bank communities.
Soil conditions also affect susceptibility. Willard and Marr
(1970) and Payne and others (1983) found vegetation on moist
soils to be most sensitive.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
The persistence of these vegetation changes in different
environments is poorly understood. Willard and Marr (1971)
estimated that hundreds of years would be required for heavily
trampled alpine tundras in the Colorado Rocky Mountains to
recover. Several studies determined use levels for which
impacts remained visible the following year. Bell and Bliss
(1973) found that 300 passes by a hiker on alpine tundra in
Olympic National Park created a path recognizable a year
later, while Payne and others (1983) showed that only eight
ORV passes were required to carve a lasting track on clay
loams and silt-clay loams at a rangeland site in Montana.
Leonard and others (1985) measured the highest rate of change
in vegetation with the first 100 to 300 passes by hikers.
Recreational developments such as marinas and ski slopes
involve major redesign of vegetation. Recreational use and
impacts are often concentrated in riparian zones, and
Settergren (1977) reviewed recreational impacts on riparian
vegetation and soils. Repeated searching for firewood has
depleted some camp sites of dead wood, and even green
wood is occasionally cut. Plant collectors, wildflower pickers,
and mushroom enthusiasts alter plant communities by
selectively removing species, and recreational boats carry
aquatic weeds from one lake to another (Johnstone and
others 1985). Wheel traffic transports spores of a root fungus
lethal to Port Orford Cedar (Zobel and others 1985), and
ORV use could further spread the disease.
Hikers and horseback riders compact soils around trails
and campsites. Dotzenko and others (1967) measured a 55
percent increase in soil bulk density and a decrease in organic
matter content in coarse-textured soils at campgrounds of
Rocky Mountain National Park, and Monti and Mackintosh
(1979) described physical changes in soils due to trampling.
Weaver and Dale (1978) compared soil impacts caused by
hikers, trail motorcycles, and horses in meadows and forests
in Montana, and found that impact level increased in roughly
that order and was worst on stone-free soils on steep slopes.
ORVs compact soils over wide areas because drivers often
desire to “break new ground,” but heavily compacted trails
are also established by repeated traffic at popular ORV
locations. Where vegetation is already sparse, traffic can
destroy the protective armoring of coarse particles and
cryptogam communities. Surface soils loosened by shear are
more erodible, and mobilized sediment may plug soil pores
(Eckert and others 1979). Motorcycles and trucks impact
soils in similar ways (Eckert and others 1979).
Some work has identified characteristics of resilient sites
to allow better planning of focal use areas. Marion and Merriam
(1985) found that soil impacts in the Boundary Waters Canoe
Area were greatest on soils with low bulk density and low
organic matter content. Summer (1980) related site sensitivity
in Rocky Mountain National Park to geomorphic classification
and showed that outcrops, talus slopes, river terraces, and
moraine tops were most resistant to change.
At sites where vegetation is depleted by trampling, soil
properties change in response. Moisture-holding capacity
decreases and bulk density increases because organic matter
is no longer cycled into the soil (Dotzenko and others 1967).
51
These changes further decrease vegetation cover and alter
plant communities to favor disturbance-resistant plants.
Foot and vehicle traffic engrave depressions in hillslopes by
repeated use, and tracks can form conduits for overland flow
where they do not follow hillslope contours. Channelized flow
is particularly pronounced along ORV tracks and where paths
cut between trail switchbacks. Improperly designed and located
trails can divert low-order channels onto their own surfaces.
Careless use of fuels and soaps by campers introduces
chemicals into the environment, and recreational vehicles
can contribute fuel, oil, combustion products, and sewage.
Littering also introduces hazardous wastes. Cryer and others
(1987) measured the litter produced by anglers at several
sites in Great Britain and found that ingestion of discarded
fishing weights and lead shot was the major cause of mortality
among British mute swans.
Angling and hunting directly alter biological communities
by targeting particular species, while management to support
these activities preferentially cultivates the target species.
Either effect can alter community structure. Cultivated strains
of game fish represent select elements of the gene pool and
may reduce genetic diversity in wild populations. Exotic
fish and game species have been widely introduced. In some
cases, substantial effort is required to protect native species
from this additional competition. Recreational activities are
also associated with increased bacteria concentrations in
surface waters (e.g., Skinner and others 1974), and may be
partially responsible for the rapid spread of Giardia.
Although most fishing within watersheds is for recreation
or subsistence, anadromous fish originating from forested
watersheds also support a large commercial fishery that strongly
affects fish populations. Hatcheries have been developed to
support both recreational and commercial fisheries. Hatchery
stock intermix with wild fish, decreasing the genetic diversity
between river systems and destroying specialized strains.
Planted fish may spread diseases to wild populations.

Table 7—Potential direct effects of selected land-use activities on watershed properties.

Auxiliary uses are also noted.

Auxiliary Vege- Topo- Chem-

use tation Soil graphy icals Other
Activity BCILRV CDP DS CFMNS INR FHPW

Construction ...LR. C.P DS CFMNS I.R F...

Auxiliary use
Impoundments .C..R. CDP .. CF.N. I.. F..W
B Burning
Channelization .C..R. CD. .. CFMN. ... ....
C Construction
Road use and maintenance .C.L.. C.. .S CFMN. I.. ..P.
I Impoundments
Vegetation conversion B..... C.P DS ..M.. INR F.P.
L Channelization
Burning ...... CDP .S ..M.. ..R FH..
R Road use/maintenance
Water development V Vegetation conversion
Transbasin imports .CILR. .D. .. ...N. I.. F.PW
Groundwater ...... ... .. ..... I.. ...W Vegetation
Timber management C Community composition
Logging and yarding .C..R. CDP DS C.MN. I.R F... D Disturbance frequency
Planting and regeneration B....V ... .. ..... ... .... P Pattern of communities
Pest and brush control B..... C.. .. ..... I.. F.P.
Fire control BC..R. CDP .. ..M.. ... .... Soils
Range use - grazing B.I.RV CDP .S C.M.. .N. F.PW D Disruption of horizons
S Altered soil structure
Mining
Open pit mining .C..R. C.P D. CFMNS I.R F...
Underground mining .C..R. ... .F... I.. .... Topography
Placer gold and gravel .C.LR. CDP D. CFMNS I.R F... C Channel/bank morphology
Tailings storage ....R. CDP DS CFMNS I.. F... F Emplacement of fill
Mine reclamation ....RV ... .S CFMN. ... .... M Altered microtopography
N Altered channel network
Agriculture S Oversteepening of slopes
Tillage and cropping ....RV .D. .S ..MN. ..R F.P.
Irrigation ..ILR. ... .. ...N. I.. ...W
Insect and weed control B..... C.. .S ..M.. I.. F.P. Chemicals
I Non-nutrient chemical input
Urbanization and power
N Introduction of nutrients
Habitation .CILR. ..P .S ..M.. IN. FHPW
R Removal of nutrients/organics
Industry .CILR. ..P .S ..MN. I.. .H.W
Power plants .CILR. ... .. ..... I.. .H.W
Other
Recreation and fishing F Faunal introduction/removal
ORVs ....R. C.. .S C.MN. I.. .... H Introduction of heat
Trails .C.... C.. .S C.MN. ... .... P Introduction of pathogens
Camping .C..R. C.. .S C.M.. IN. ..P. W Import/removal of water
Fishing and hunting ....R. ... .. ..... I.. F...

52 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Recreational use commonly overlaps with timber
management activities. Logged areas attract hunters because
they provide habitat for some game species. In addition,
access provided by logging roads increases recreational use
in adjacent undisturbed areas, and logging may be accompanied
by recreational developments such as campgrounds. Mountain
bikes and ORVs often use logging roads.
General Patterns of Change
Most land use can directly affect only vegetation, soils, to-
pography, chemicals, fauna, and water, and it can affect these
environmental parameters in only a limited number of ways.
Table 7 lists changes resulting directly from specified activi-
ties. Only habitation, industrial activities, and fire are shown
to directly add heat, for example, although many other activi-
ties influence heat distribution by modifying channel mor-
phology, microclimate, and shading. Similarly, only activi-
ties that import water affect the amount of water entering a
basin, although many others influence the amount of runoff a
catchment generates. Each of the activities listed is associ-
ated with other types of land use, and these associations are
indicated in the “auxiliary use” category. Thus, most activi-
ties require road use, which itself requires construction ac-
tivities and channelization of road-related drainage.
Direct effects of land-use activities on vegetation include
changes in community composition, disturbance frequencies,
and the pattern of vegetation communities. These can result
from intentional removal of vegetation, as in the case of
logging or construction, or selective pressure from grazing
or trampling, or addition of desired species during agricultural
or silvicultural vegetation conversions. Land-use activities
can also directly alter soil characteristics. Many activities
modify soil structure through compaction or loosening, and
others disrupt soil profiles by scarification or excavation.
Common topographic changes on hillslopes include
oversteepening of slopes, accumulation of unconsolidated
materials, and alteration of soil surface roughness. Land use
also reroutes channel networks and alters channel morphology.
Changes to chemical parameters usually involve import or
export of chemicals or nutrients. This may be by removal of
watershed products or by addition of waste materials or
management aids. Other direct changes include introduction
of pathogens or heat, and input or removal of fauna or water.
One environmental parameter usually influences many
others, and the effects of changes in environmental parameters
on others are outlined in table 8. These constitute indirect
effects of land use on environmental parameters, but are
direct effects of changes in environmental parameters. For
example, compaction of soils usually excludes some plant
species from a site, and thus alters the vegetation community
present. This change, in turn, can trigger changes in nutrient
inputs, fauna, and disturbance frequency. Similarly, soil

Table 8—Interactions between altered environmental parameters; letters indicate effect of

parameter listed in row on parameter listed in column.

Vege- Topo- Chem-

tation Soil graphy icals Other
CDP DS CFMNS INR FHPW

Vegetation
C Altered community composition .DP DS ..M.. .N. F...
D Altered disturbance frequency C.. DS ..M.. .NR F...
P Altered pattern of communities .D. .. ..... ... F...
Soils
D Disruption of horizons C.. .S ..M.. ... F...
S Altered soil structure C.. .. ..... ... F...
Topography
C Altered channel or bank morphology CD. .. ..M.S ... F...
F Emplacement of fill C.. DS C.M.S ... ....
M Altered microtopography ... .. ..... ... ....
N Altered channel network .DP .. ..... ... F...
S Oversteepening of slopes .D. .. ..... ... ....
Chemicals and nutrients
I Import of non-nutrient chemicals C.. .. ..... ... F...
N Introduction of nutrients C.. .. ..... ... ....
R Removal of nutrients or organics C.. .. ..... ... ....
Other effects
F Introduction or removal of fauna CD. DS C.M.. .N. ..P.
H Introduction of heat C.. .. ..... ... F...
P Introduction of pathogens CD. .. ..... ... F...
W Import or removal of water CD. .. C.... I.. F.P.

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 53

characteristics are themselves influenced by the vegetation
community present, and faunal changes can affect most
other environmental characteristics.
Cumulative Effects of Land Use
on Environmental Parameters
Cumulative impacts on particular resources and values are
considered in later chapters. However, environmental param-
eters can undergo cumulative change without reference to
resources or values, and impacts that result from a cumulative
change to an environmental parameter can also be considered
CWEs. Cumulative changes to environmental parameters can
be generated by several types of land-use interactions.
Effects of land use can be cumulative if changes caused
by an activity persist long enough to interact with subsequent
activities. Repeated trampling of a site over a short period
can cause cumulative compaction, because most soils require
long periods to recover from compaction. Decreased
vegetation cover or altered plant communities at compacted
sites thus are cumulative effects of multiple vehicle passes.
Progressive changes in activity characteristics can also lead
to cumulative effects, as can changes caused by a single
activity that continues for a long duration. For example,
impervious area increases through time in urban areas because
development density increases as urban centers age.
Some effects can accumulate from one activity cycle to
the next. For example, seasonally grazed pastures may show
cumulative changes in species composition, and multi-cycle
clearcutting can cumulatively deplete wood inputs to the
soil. Changes in soil fertility may also accumulate through
cutting or cropping cycles as unreplenished nutrients are
exported with the harvest, and repeated mining of gravels can
cumulatively incise a channel if extraction rates are higher
than replenishment rates.
A single type of activity may progressively alter an
expanding area, as when effects of urbanization accumulate
in a watershed as population centers grow: impervious area
increases through time, so areally averaged infiltration rates
progressively decrease. Progressive enlargement of an open-
pit mine can cumulatively increase the impacts on
environmental parameters, and logging has progressively
reduced the area of old-growth forests since European
settlement of North America.
Many activities are performed in sequence, and effects
may accumulate if responses to early activities are persistent.
Logging commonly precedes agriculture or grazing, and
each use produces environmental changes that can reinforce
one another. Removal of trees exports nutrients from
watersheds, as does subsequent cropping, so nutrient capital
can be cumulatively depleted if recovery is slow.
Activities occurring at the same time in different parts of
a watershed can also cumulatively influence environmental
54
parameters. Road use, grazing, logging, and mining all induce
soil compaction, and together they cumulatively decrease
the areally averaged infiltration rate in a watershed.
Research Needs for
Understanding Changes in
Environmental Parameters
Science is an ongoing process of information refinement, so
research rarely provides complete answers to questions. From
a management perspective, a research project is successfully
completed when it provides the detail of understanding and
information necessary to improve management procedures as
much as desired. Most of the topics described in this chapter
could profit by further work, but some will provide greater
returns in the form of practical applications, and some require
more work because they are less well understood.
Understanding Environmental
Parameters
Fundamental to all management applications is develop-
ment of a basic understanding of how natural systems func-
tion. Effects caused by land use are important only if they
differ in character or magnitude from those occurring natu-
rally, so natural patterns of disturbance must be understood.
Little is known of long-term patterns of vegetation distur-
bance and recovery in natural ecosystems. We need to know
more about the roles played by disturbance and by adjacent
and interfingering communities in the maintenance of veg-
etation communities. Qualities lending resilience to biologi-
cal systems must also be better understood, as well as depen-
dencies that make some species more sensitive to change.
The effects of interactions between disturbances are very
poorly understood.
Soil distribution is comprehensively mapped, yet little is
known of soil formation rates, or of how rapidly soil
characteristics change in response to environmental change.
Disturbance is important in maintaining soil character, yet
its role and frequency are poorly understood. Compaction is
a common soil impact and is induced by most land uses, so a
method of predicting a soil’s susceptibility to compaction and
its recovery time would be an extremely useful management
tool. Long-term effects of fire on soils are poorly understood,
and recovery mechanisms and persistence of fire effects need to
be documented.
Watershed chemistry is more poorly described than
vegetation or soils. Rates and mechanisms of natural nutrient
production need to be better understood before the nature
and duration of human-induced impacts can be predicted.
Reactions of vegetation to altered nutrient inputs require
measurement.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Understanding the Effects of Land Use
The direct effects of many types of land use are known, but
their distribution and persistence have rarely been adequately
described. Instead, many studies focus on the implications
of land use for watershed processes. Clearcut area is corre-
lated to changes in runoff, for example, without exploring
reasons for the changes. As a result, mechanisms of change
are not fully understood, so study results cannot be applied
to other sites. Long-term changes due to land use also must
be defined to predict future effects, and rates and mecha-
nisms of recovery need to be measured.
A very basic research need for most land uses is a simple
inventory of what practices are being used in what areas,
how these uses are changing, and how closely practices
adhere to guidelines. Road construction is heavily regulated,
for example, but little is known of how closely roads actually
conform to guidelines. Only where violations result in
immediate, large-scale damage, as was the case with the
Highway 101 bypass project in Redwood National Park, are
shortcomings recognized. Other problems are more subtle.
For example, Piehl and others (1988b) found that most of the
culverts they inventoried in the Oregon Cascades were
undersized.
Many road-related changes result from chemicals
introduced by construction or traffic. The distribution and
persistence of these chemicals is poorly understood, as are
their effects on surrounding vegetation communities. Effects
have been measured, but they are not yet understood well
enough to allow prediction.
Both State and Federal timber-management guidelines
leave many decisions to land managers. The frequency,
nature, and effects of these decisions need to be determined,
and unintentional or overlooked variances from established
guidelines also require inventory. Eckerberg (1988)
demonstrated that observance of guidelines in Sweden was
biased: measures that did not restrict logging were most
carefully observed, while those designed to protect other
interests at the expense of logging efficiency were often
overlooked.
Particular aspects of timber management that require further
study include measurement of the effects of different cutting
patterns and strategies on the overall community composition
of a watershed. Resulting patterns must be compared with
natural disturbance patterns and ecosystem structures to
understand long-term changes that might result. In particular,
the effects of management and cutting strategies on disease
outbreaks, insect infestations, and genetic diversity require
more complete understanding, and long-term effects of
burning on community structure, diversity, pattern, and soil
chemistry need to be determined. Chemical inputs and exports
from forest practices are poorly defined for most areas and
require basic descriptive work. The effects of previous timber-
management practices on modern vegetation patterns and
channel morphology must be determined. Splash dams and
log drives significantly altered channel morphology in many
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
small channels, and most logging today is in second-growth
forests. Many systems of concern today are no longer in
their natural state and may still be adjusting to changes of
the past century.
Not much is known of the pattern of grazing plan
implementation on public and private lands, and the
correspondence between plans and implementations needs
to be inventoried. One study of management-related
differences showed that range condition in Utah was positively
correlated with expenditure (Loring and Workman 1987).
Although the effects of grazing strategies and management
efforts on vegetation type are usually well-described, long-
term implications are more poorly understood. Physical
changes caused by grazing also require better description.
The effects of trampling on channel morphology need to be
systematically determined and related to channel
characteristics, and recovery rates and mechanisms must
also be determined. Rates of compaction and recovery from
trampling need to be measured as a function of soil type and
texture. Grazing can affect water quality by introducing
bacteria to streams, but the conditions under which these
inputs might be detrimental are not well known.
The effects of early mining activity are poorly documented
even where they affect ongoing use, and the extent to which
today’s channels reflect mining disturbance is usually
unknown. Weathering of exposed mine spoils can be a
persistent source of chemical pollutants, but factors controlling
it are poorly understood. Short- and long-term effects of in-
channel and floodplain mining are not known, and the extent
to which mining-related disruption destabilizes channel beds
and introduces fine sediments needs to be determined.
The effects of introduced plant species on California’s
original vegetation communities are poorly documented, and
some communities and physical systems may still be
responding to vegetation changes that occurred a century
ago. Agricultural chemicals cause many of the same problems
as chemicals introduced by roads and timber management,
and also require evaluation for their persistence and role in
the environment.
Urban effects of particular note are introductions of toxic
waste and atmospheric effluent, and altered water inputs by
transbasin imports and groundwater extraction. The extent of
each of these and their effects on vegetation need to be
determined. Associated uses, such as power development,
impoundments, water development, flood control, and
navigation, strongly affect channel morphology. The direct
influence of these uses is well understood, and their effects on
watershed processes will be considered in the following section.
Recreation is a rapidly growing aspect of land use, with
more people making use of a diminishing recreational land
base. Patterns of recreational use and factors attracting
recreationists to particular sites must be identified to understand
future interactions between recreation and other uses.
55
Chapter 6
Effects of Land Use and Environmental Change on
Watershed Processes

M ost land-use activities alter environmental param-eters

in a limited number of ways: they change the charac-
ter of vegetation and soils; they modify topography; they
import and remove chemicals, water, and fauna; and they
introduce pathogens and heat. As these environmental pa-
rameters change, processes associated with transport of wa-
ter through watersheds change in response and alter the
production and transport of water, sediment, organic matter,
chemicals, and heat. This chapter describes how changes in
environmental parameters affect generation and transport of
these watershed products.
Relation Between Watershed
Processes and CWEs
Off-site CWEs can be generated only by changes in the transport
of watershed products, because the effects of a land-use activity
can influence a remote site only if something is transported to
that site. Changes in the production or transport of watershed
products can also cause impacts at the site of land use.

Table 9—Effects of altered environmental parameters on watershed processes.

Runoff Sediment Organics Chemicals Heat

PIEMHCY AHCBY ADHRCBY AHCBY AW

Vegetation
C Altered community composition P.E.H.. .HC.. A.H..B. A.... AW
F Altered disturbance frequency ....... .H... A....B. ..... ..
P Altered pattern of communities ....... ..... ....... ..... ..
Soils
D Disruption of horizons ....... A.... ....... A.... ..
S Altered soil structure .I..... .H... ....... ..... ..
Topography
C Altered channel/bank morphology .....C. ..C.. ....C.. ..... .W
F Emplacement of fill ....... A..B. ....... ..... ..
M Altered microtopography ....H.. .H... ..H.... ..... ..
N Altered channel network .....C. ..C.. ....C.. ..... .W
S Oversteepening of slopes ....H.. .H... ..H.... ..... ..
Chemicals and nutrients
I Import of non-nutrient chemicals ....... ..... ....... A..B. ..
N Introduction of nutrients ....... ..... ....... A..B. ..
R Removal of nutrients or organics ....... ..... A...... ..... ..
Other effects
F Introduction or removal of fauna ....... .HC.. A...... ..... ..
H Introduction of heat ....... ..... ....... ..... AW
P Introduction of pathogens ....... ..... ....... ..... ..
W Import or removal of water P....CY ..... ....... ...B. .W

Runoff Organic material Heat

P Production process
I Infiltration
E Evapotranspiration
M Soil moisture
H Hillslope hydrograph
C Channel hydrograph
Y Annual water yield
Sediment
A Amount and character on hills
H Hill erosion process and rate
C Channel erosion process and rate
B Amount and character in channel
Y Sediment yield and character
A Amount and character on hills A Air temperature
D Decay rate on hillslopes W Water temperature
H Transport rate on hillslopes
R Decay rate in channel
C Transport rate in channel
B Amount and character in channel
Y Volume and character exported
Chemicals
A Soil chemistry
H Transport on hillslopes
C Transport in channel
B River chemistry
Y Volume and character exported

56 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Whenever multiple activities or activities at multiple sites
produce similar or complementary changes to environmental
parameters or watershed processes, impacts resulting from
those changes can be considered cumulative. Table 9 indicates
the direct effects that altered environmental parameters can
have on watershed products, and table 10 shows the effects
of interactions between watershed products. These tables can
be used in conjunction with table 7 and table 8 to recognize
land-use interactions that might generate CWEs. Cumulative
effects on an environmental parameter can occur when two
land-use activities affect the same parameter (table 7), or
affect other environmental parameters that influence the same
parameter (table 8). Similarly, changes can be considered
cumulative if two altered parameters affect the same watershed
product (table 9), or affect other watershed products that, in
turn, affect the same product (table 10). Chapter 7 describes
the effects of these cumulative changes on particular resources
and values.
The tables are not useful for identifying particular effects
of land-use activities, because each of the activities listed in
table 7 is directly or indirectly capable of altering each of the
product attributes or processes listed in table 10; this is
evident when the effects of a particular activity are traced
through the sequence of tables. However, the tables can be
used to make clear the pathways of influence that can result
in environmental change, and if information for a particular
site indicates that some influences are inconsequential, then
the tables can be used to narrow the field of likely effects.
Effects of Environmental Change
on Hydrology
Runoff volume, its mode and timing of production, and its
rate of transport through a channel system all affect both the
rate of water delivery to any point and its ability to transport
other watershed products. Changes in any of these character-
istics affect downstream conditions. Published research that
describes the effects of logging and roads on hydrology is
listed in table 11.
Water Input
Canopy cover influences how much precipitation the ground
receives. Snow and rain trapped in foliage evaporates rap-
idly because it is exposed to wind, so less snow and rain

Table 10—Interactions between watershed processes. Symbols are explained in the left column.

Runoff Sediment Organics Chemicals Heat

PIEMHCY AHCBY ADHRCBY AHCBY AW

Runoff
P Production process ....H.. .H... ....... ..... .W
I Infiltration P..M... ..... ....... A.... ..
E Evapotranspiration ...M.CY ..... ....... A.... A.
M Soil moisture .IE.... .H... .D..... A.... ..
H Hillslope hydrograph .I...C. .H... ..H.... .H... ..
C Channel hydrograph ....... ..CBY ....C.. ..C.. .W
Y Annual water yield ....... ..... ....... ..... ..
Sediment
A Amount and character on hills ..E.H.. .H... ....... ..... ..
H Hill erosion process and rate ....... A.CB. ..H.... .H... ..
C Channel erosion process and rate ....... .H.BY ..H.... ..C.. ..
B Amount and character in channel .....C. ..C.. ....... ...B. .W
Y Sediment yield and character ....... ..... ....... ..... ..
Organic material
A Amount and character on hills .IE.H.. .H... .DH.... A.... ..
D Decay rate on hillslopes ....... ..... A...... A.... ..
H Transport rate on hillslopes ....... ..... A....B. .H... ..
R Decay rate in channel ....... ..... .....B. ...B. ..
C Transport rate in channel ....... ..... .....BY ..C.. ..
B Amount and character in channel .....C. ..CB. ...RC.Y ..C.. .W
Y Volume and character exported ....... ..... ....... ..... ..
Chemicals
A Soil chemistry .I..... .H... .D..... .H... ..
H Transport on hillslopes ....... ..... ....... A..B. ..
C Transport in channel ....... ..... ....... ....Y ..
B River chemistry ....... ..... ...R... ..C.Y ..
Y Volume and character exported ....... ..... ....... ..... ..
Heat
A Air temperature P.E.... ..... .D..... A.... .W
W Water temperature ....... ..... ...R... ...B. ..

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 57

 Table 11—Studies of the hydrologic effects of roads and timber management. Symbols are explained below.

Treatment Measurements
Reference Location Approach Scale F R T CEFILMPSTWY

Anderson and Hobba 1959 OR A B T P

Andrus and others 1988 OR D C T W
Aubertin and Patric 1974 WV T B T Y
Austin and Baisinger 1955 OR TS P F M
Beasley and Granillo 1988 AR T M T Y
Beaton 1959 BC AS P F F
Berndt and Swank 1970 OR AT C T Y
Berris and Harr 1987 OR T P T ST Y
Beschta 1990 - O - F F L P Y
Bethlahmy 1962 OR T P T M
Bethlahmy 1967 ID E P T Y
Blackburn and others 1986 TX T M T Y
Bosch and Hewlett 1982 - O - T E Y
Bren and Leitch 1985 AU TM CP R P T Y
Bryant 1980 AL T P T W
Carlson and others 1990 OR S P T W
Caspary 1990 WG T C T E Y
Chanasyk and Verschuren 1980 BC M HC T E M Y
Cheng 1989 BC T B T P Y
Cheng and others 1975 BC T B T P T
Christner and Harr 1982 OR A M T PS
Courtney 1981 - O - T E I
Cullen and others 1991 MT S P R T C M
Debano 1981 - O - F F
DeByle and Packer 1972 MT T P F T Y
Dickerson 1976 MS E P R T C F
Dietterick and Lynch 1989 PA T B T P Y
Dyrness 1965 OR S P R T C
Dyrness 1976 OR T P F F
Ffolliott and others 1989 AZ,NM O - T S Y
Froehlich 1979 OR A P R T C
Froehlich and others 1985 ID T P R T C
Geist and others 1989 OR A P R T C
Gent and Ballard 1985 NC T P R T C
Gray and Megahan 1981 ID T P F R T M
Greacen and Sands 1980 - O - R T C
Green and Stuart 1985 GA E P R T C
Harr 1980 OR T B T Y
Harr 1986 OR T MC T PS
Harr and others 1975 OR T B R T P T Y
Harr and others 1979 OR T M R T C L P Y
Harr and others 1982 OR T B T L P Y
Harr and McCorison 1979 OR T CB T PST
Harris 1977 OR T B T P Y
Hatchell and others 1970 SC,VA AES P R T C
Heede and King 1990 AZ T P T Y
Hewlett and Helvey 1970 NC T C T P T
Hicks and others 1991 OR T M F T L
Hillman and Verschuren 1988 - M H T M
Hogan 1987 BC D PB T W
Hornbeck 1973 NH T B T PST
Hornbeck and others 1970 NH T B T L S Y
Johnson and Beschta 1980 OR S P R T
Keppeler and Ziemer 1990 CA T B R T L Y
King 1989 ID ST M R T PS Y
King and Tennyson 1984 ID T M R L P T Y
Klock and Helvey 1976 WA T B F MP
Lockaby and Vidrine 1984 LA S P R T C
Mahacek-King and Shelton 1987 CA M C T P Y
Martin and Tinney 1962 OR A C T Y
McCarthy and Stone 1991 FL E M T M
McDade and others 1990 OR,WA S P T W
McNabb and others 1989 OR T P F F M
Megahan 1972 ID T P R P
Megahan 1983 ID T P F R T MPS

58 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Table 11—(Continued)

Treatment Measurements
Reference Location Approach Scale F R T CEFILMPSTWY

Moehring and Rawls 1970 AR S P R T C

Patric and Gorman 1978 WV AD P R T C F
Patric and Reinhart 1971 WV T C T F L P Y
Pierce and others 1958 NG S P T F
Potts 1984 MT A C L P T
Potts and others 1989 - M P F
Reinhart 1964 WV T CP R T F PS
Rice and Wallis 1962 CA T B T Y
Riekerk 1989 FL T B T C F MP
Rothacher 1963 OR T P T I
Rothacher 1970 OR T B R T E L Y
Rothacher 1973 OR T B R T P
Savage 1974 L E P F F
Schmidt and Troendle 1989 CO T P T
Scott 1956 CA S P F F
Scott and Burgy 1956 L E P F F
Sidle and Drlica 1981 OR S P R T C
Snider and Miller 1985 OR D P R T C F
Springer and Coltharp 1980 KY T B R P T
Steinbrenner and Gessel 1955 WA D P R T C
Stransky 1981 TX T P T C
Striffler 1959 MI ST P T F
Swank and Helvey 1970 NC T B T Y
Tarrant 1956 WA,OR S P F T F
Toews and Moore 1982 BC T B T W
Troendle 1970 WV T B T L
Troendle 1988 CO T BP T E M S Y
Ursic and Esher 1988 MS R C T Y
Van Sickle and Gregory 1990 OR M P T W
Wood and others 1989 LA E P T Y
Wright and others 1990 CA T B R T P T Y
Ziemer 1964 CA A P T E M
Ziemer 1979 - O - T E
Ziemer 1981 CA T B R T P T Y

Location:
AL Alabama FL Florida MS Mississippi PA Pennsylvania
AR Arkansas GA Georgia MT Montana SC South Carolina
Au Australia ID Idaho NC North Carolina TX Texas
AZ Arizona KY Kentucky NG New England VA Virginia
BC British Columbia L Laboratory NH New Hampshire WA Washington
CA California LA Louisiana NM New Mexico WG West Germany
CO Colorado MI Michigan OR Oregon WV West Virginia
Approach or method: Scale:
A Reconstructs past using air photos or B Paired basins or sites
records, or by sampling age distribution C Case study
D Based on descriptive measurements H Hypothetical or general case
E Experimental M Multiple basins or sites
M Modeling or calculated results P Process study
O General review of topic
R Reasoned from basic principles Treatment:
S Survey of multiple sites F Fire
T Long-term temporal monitoring study R Roads
T Timber management
Measurement:
C Compaction of soil P Peak flow
E Evapotranspiration S Snow melt or accumulation
F Infiltration into soil T Timing of hydrograph peak during storm
I Interception loss W Woody debris
L Low-flow characteristics Y Water yield
M Soil moisture

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 59

reaches the ground in forested areas than in adjacent clear-
ings. Berris and Harr (1987) showed that snow accumulation
and its water equivalent was higher in clearings than in
adjacent forest in the Oregon Cascades, and Rothacher (1963)
reported loss of 18 to 24 percent of rainfall to evaporation
from foliage in the same area. On the other hand, fog can
condense on leaf surfaces and contribute some precipitation
by fog drip. Courtney (1981) reviewed studies of intercep-
tion loss and noted that merely measuring the rates is not
sufficient; studies can be interpreted only if the processes
that create the effect are understood.
Some land-use activities modify climate. Precipitation or
cloudiness may be increased locally by intentional cloud
seeding or seeding by atmospheric pollutants. Kahan (1972),
for example, quoted increases of 10 to 30 percent in local
precipitation from cloud-seeding programs. Local increases
may be balanced by decreases downwind where atmospheric
moisture content is reduced. Increased cloud or fog cover
may increase runoff by decreasing evapotranspiration rates.
Broader changes may be caused by global increases in
atmospheric CO2 and other gases. Whether atmospheric
alterations will increase or decrease precipitation at a site
depends on the site’s location and the type and extent of
atmospheric change. Climatic models have been developed
to predict the likely effects of atmospheric trends, but model
validity is difficult to assess, and spatial resolution is sufficient
only for regional projections. Gibbs and Hoffman (1987)
reviewed available climatic models and discussed their
shortcomings, strengths, and assumptions. Rathjens (1991)
described the consensus he perceives among climate change
researchers: there has already been a 0.5 °C temperature
increase over the past 100 years; there will likely be a 1.5-4.5
°C increase by 2035; and this will be accompanied by a 0.2-
to 0.5-m rise in sea level. He noted that the magnitude of the
temperature increase may be lessened due to increased albedo
from increased cloudiness, but that the increased cloudiness
might also enhance a temperature increase by altering the
atmospheric water vapor content. In any case, increased
cloudiness would in itself markedly change local and global
precipitation patterns, irrespective of temperature effects.
Changes in precipitation timing, mode, and intensity also
affect watershed hydrology. Increased dry-season rainfall is
likely to have more effect on vegetation than equivalent
increases during the wet season, and a shift from rain to
snow alters annual hydrographs. Flood peaks will respond if
rain intensities increase. Lettenmaier and Gan (1990) used
global circulation models to predict hydrologic changes in
California due to global warming, and suggested that an
increased proportion of precipitation falling as rain would be
a more important influence than an increase in the amount of
precipitation.
Irrigation and domestic use can augment flow downstream
of developments by importing water from other areas. Glancy
and Whitney (1989) described transformation of Las Vegas
Wash from an ephemeral channel to a perennial stream with an
average discharge of 4 m3/s as the population of Las Vegas grew.
60
Runoff Generation
Changes in vegetation density and age structure affect rates
of evapotranspiration, so altered vegetation usually changes
runoff volume and timing. Ziemer (1979) reviewed research
on evapotranspiration processes. Decreased evapotranspira-
tion increases average soil moisture (Bethlahmy 1962), raises
dry-season water tables, and augments dry-season baseflows.
These changes can increase storm peaks early and late in the
wet season, but mid-season peaks are rarely affected be-
cause soil moisture is usually high at this time even before
disturbance. Rothacher (1973) demonstrated this pattern in
the Oregon Cascades, but Hewlett and Helvey (1970) showed
that altered soil moisture may affect peaks more consistently
in areas where precipitation is distributed throughout the
year. Knox and others (1975) found three-fold increases of
peak flows after forests were converted to cropland in the
northern Midwest.
Vegetation may be intentionally modified to increase
water yield, but altered yields more commonly are unintended
by-products of other uses. Bosch and Hewlett (1982) reviewed
studies of the hydrologic effects of vegetation management
and showed a pattern of increased water yield after
clearcutting. Harr and others (1979), for example, measured
a 43 percent increase for 5 years after a clearcut in the
Oregon Coast Range. Martin and Tinney (1962), however,
found no change in runoff in a 55 mi2 basin nearby and
suggested that regrowth can be rapid enough to mask the
hydrologic effects of logging if the cutting rate is very low.
Swank and Helvey (1970) estimated that recovery of runoff
rates from a North Carolina clearcut will require 35 years,
while Ziemer (1964) documented recovery after about 16
years in the Sierra Nevada. Evapotranspiration rates are also
affected by forest declines from air pollution. Caspary (1990)
found increased water yield in pollution-stressed forests even
before foliar damage was evident, and attributed this change
to the physiological effects of acid soils on fine roots.
Snow melts more quickly in clearings than under forest
canopies because air circulation and solar radiation are higher.
Hornbeck and others (1970) documented an earlier snowmelt
season in clearcuts at the Hubbard Brook Experimental Forest
in New Hampshire, and Berris and Harr (1987) extended the
argument to predict more rapid runoff during particular melt
events. In combination with higher snow accumulations in
clearings, the increased melt rates can generate higher peak
flows. This effect is enhanced if snow is melted during rain
storms, and clearcutting in the transitional snow zone may
increase flood peaks during rain-on-snow events (Christner
and Harr 1982, Harr 1986). Buttle and Xu (1988) demonstrated
that urbanization also caused earlier snowmelt and augmented
discharge peaks from rain-on-snow events.
Processes of runoff generation and evapotranspiration
loss are relatively well described, so models can be constructed
to predict the effects of altered forest vegetation on basin
hydrology. Chanasyk and Verschuren (1980) developed a
mathematical model to predict the effect of clearcutting on
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
soil moisture and runoff, and Hillman and Verschuren (1988)
used a version of this model to evaluate the influence of
cutting pattern on soil moisture, but cautioned that the model
is not capable of handling macropore flow or spatial variations
in soil properties.
Riparian vegetation may decrease baseflows by transpiring
stream water. Orme and Bailey (1971) documented increased
baseflow after removal of riparian plants and conversion of
chaparral vegetation to grassland in southern California,
although accompanying morphological changes were
catastrophic. Van Hylckama (1974) measured water use by
tamarisk along floodplains of the Gila River in Arizona and
showed evapotranspiration losses of more than 200 cm/yr in
dense thickets.
Overland flow is produced by rainfall and snowmelt to the
extent that the rate of water input exceeds infiltration into the
soil, and infiltration rates are controlled by soil properties
and vegetation. Infiltrated water is held in the soil mantle
when soils are dry, but wet soils can transmit water downward.
Compaction of soil decreases pore space and collapses conduits
between pores, reducing soil porosity and permeability and
increasing rates of overland flow. Removal of topsoil can
also cause overland flow. Sponge-like humic layers may
store enough water that through-flow is slowed to rates that
less permeable horizons can absorb. Without these surface
horizons, excess rainfall may run off before infiltrating.
Unsurfaced roads and construction sites are often highly
compacted and readily generate runoff. Even briefly used skid
roads show impaired infiltration (e.g., Dickerson 1976). Johnson
and Beschta (1980) measured reduced infiltration capacities
on skid trails, cable paths, windrows, and burned surfaces at
sites in western Oregon, but showed that infiltration on clearcuts
was unchanged in the absence of these other activities.
Trampling by animals or people also compacts the soil
and alters its hydrologic properties. Gifford and Hawkins
(1978) reviewed the impacts of grazing on infiltration and
compiled results of published studies. Heavy grazing was
generally found to reduce infiltration rates on porous soils
by about 50 percent, while light and moderate grazing
decreased rates to about 75 percent of their original values.
Blackburn and others (1980) reviewed hydrologic effects of
different grazing strategies and described studies in Texas
that indicated the site-specific nature of results. Thurow and
others (1988) found average stocking rate to be more important
than grazing strategy in influencing infiltration rates on the
Edwards Plateau in Texas, and Warren and others (1986b)
found that recovery of the range in that area was controlled
more by the length of rest period than by short-term stocking
density for a given time-averaged stocking rate. Packer (1953)
artificially trampled plots with a steel “hoof” to measure the
effects of stocking rate on infiltration.
Several studies have examined basin-scale runoff from
grazed lands. Higgins and others (1989) found no change in
runoff for several grazing intensities in eastern Oregon,
while Lusby (1970) demonstrated a 40 percent increase in
runoff from grazed land in western Colorado. Results are
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
expected to be dependent on climate, soil type, stocking
density, and vegetation type.
Compaction can also be severe on recreation sites. Eckert
and others (1979) measured changes in infiltration capacity
caused by ORV traffic on desert soils of southern Nevada,
and Monti and Mackintosh (1979) measured decreased
infiltration capacities at camp sites in northwest Ontario.
Dotzenko and others (1967) documented decreased soil
moisture in campgrounds in Rocky Mountains National Park.
Replacement of natural surfaces by impermeable material
in urban areas allows a high runoff ratio and increases both
peak flows and total runoff volume. Owe (1985), for example,
showed a 51 percent increase in annual runoff after
development of an area in Pennsylvania. Hollis (1975) designed
a procedure for predicting changes in flow peaks as a function
of flood recurrence interval and percent of the basin surface
that is impervious, and Stankowski (1972) developed equations
to estimate the percent of an area impervious as a function of
population density. Driver and Troutman (1989) produced
regression equations to predict runoff characteristics for urban
areas throughout the United States.
Physical disturbance by plowing or scarification may
increase soil permeability and infiltration capacity, and the
accompanying increases in soil surface roughness can slow
overland flow and allow it more time to infiltrate. Increased
infiltration decreases runoff peaks, increases low-flow
discharge, and decreases runoff volume by providing more
water for evapotranspiration. However, sediment carried by
rainsplash and surface erosion can plug surface pores of
some exposed soils, and infiltration rates at these sites can
be reduced to less than pre-disturbance levels. Reconstruction
of soils on mine spoils and reclaimed mine land can either
enhance or decrease infiltration, depending on the severity
of compaction and the type of material. Gilley and others
(1977) found decreased infiltration capacities on coal mine
spoils that contained sodic sediments, because high sodium
content increased soil dispersion and promoted formation of
a surface crust that retarded infiltration.
Burning of some plants releases volatile oils that can coat
soil particles to form a water-repellent (hydrophobic) layer
that restricts infiltration. Dyrness (1976) measured a ten-
fold decrease in infiltration rates after a wildfire in the
Oregon Cascades, and McNabb and others (1989) found a
15 percent decrease after a slashburn in southeastern Oregon.
Runoff rates increase if hydrophobic effects are widespread,
and flood peaks may be increased, low-flow discharges
diminished, and available moisture decreased. Debano (1981)
reviewed research on hydrophobic soils and described
characteristics of soils susceptible to water repellency.
Reservoirs alter runoff volumes by increasing the area of
water surface susceptible to evaporation. In addition,
impoundments held for consumptive use can remove an
appreciable proportion of a small catchment’s runoff. Milne
and Young (1989) found that stockponds in the Little Colorado
watershed retain up to 7 percent of the basin’s runoff during
dry years.
61
 Alteration of other watershed processes can also influence
rates, timing, and modes of runoff generation. Erosion often
leaves mineral soil exposed, which increases rates of overland
flow and, in turn, promotes erosion. Walker and Everett
(1987) described acceleration of snowmelt near a road in
arctic Alaska where road dust decreased albedo of the snow.
Water Transport on Hillslopes
Ground-cover vegetation slows surface runoff and decreases
peak discharges by roughening the flow path. Temple (1982)
and Petryk and Bosmajian (1975) described methods for esti-
mating vegetation roughness. Where surface flows are slowed,
water has more time to infiltrate and recharge soil moisture.
Land-use activities often affect hillslope flow by altering
topography. Roadcuts and other excavated faces can increase
overland flow by intercepting subsurface flow, and Megahan
(1972) documented the importance of this mechanism in
permeable Idaho soils. Similarly, channel incision in response
to gravel mining can lower streamside water tables (Sandecki
1989), and open-pit mines and quarries can intercept aquifers.
Grazing, tillage, yarding, and ORVs engrave depressions on
hillslopes. Where depressions are along contour they roughen
the surface, slow overland flow, and promote infiltration, but
where they are parallel to gradient, flow can collect and accelerate.
Road ditches and ruts commonly act as channels, and flow
generated on impervious road surfaces or intercepted by roadcuts
rapidly enters this expanded channel system and often is
efficiently routed to natural streams. Artificial expansion of the
drainage network is extreme in urbanized areas. Each paved
street becomes part of the network and rapidly carries runoff to
sewer systems designed for efficient conveyance. Urban peak
flows are increased not only because of the vast increase in
impermeable area, but also because of the efficiency of the
engineered drainage network. Graf (1977) measured a 50 percent
increase in drainage density after suburbanization near Iowa
City and attributed changes in runoff characteristics to the
altered channel network.
Several models have been developed to predict the effects
of urbanization on hydrographs. Arnold and others (1987)
used such a model to reconstruct pre-urbanization runoff
near Dallas and compared predicted and measured yields to
show a 12 percent volume increase from urbanization. Ng
and Marsalek (1989) used an urban runoff model developed
by the U.S. Environmental Protection Agency to predict
changes in runoff volume and flood peaks from hypothesized
land-use changes in Newfoundland.
Water Transport in Channels
Streambank and channel vegetation contribute to increased
peak stages by slowing flood flows. Burkham (1976) mea-
sured this effect for floods on the Gila River and demon-
strated differences in flood character for three conditions of
bank vegetation. Pasche and Rouve (1985) developed a model
to predict flow resistance imparted by floodplain vegetation.
Where flow is slowed by vegetation it deposits part of its
62
sediment load and thus contributes to floodplain construction,
channel aggradation, and loss of flow capacity. Graf (1978)
documented a 27 percent decrease in channel width after
colonization of Colorado Plateau streambanks by introduced
tamarisk. Encroachment of riparian vegetation is common
where dams decrease downstream peak flows. Northrup (1965),
for example, found channel capacities downstream of dams
on the Republican River to have decreased by two-thirds, and
cited vegetation encroachment as a contributing cause. En-
croachment can also be encouraged by heightened water tables
due to floodplain irrigation (Nadler and Schumm 1981).
Channel hydrographs are also influenced by channel form,
which is controlled, in part, by bank stability. Plants contribute
to bank stability by increasing soil cohesion and shielding
erodible sediment from high-velocity flow. Headward
expansion of channel networks speeds delivery of water
downstream and contributes to increased flood peaks and
runoff volumes. Channel network stability is influenced by
the resistance of unchannelled swales to erosion, and here,
too, resistance is often controlled by vegetation cover.
Developed areas are usually protected from flooding by
levees, and some reaches may be fully channelized to improve
flood conveyance and prevent channel migration. Flood flows
are no longer slowed by floodplain roughness at these sites,
and they are transmitted more rapidly through the channelized
reach. However, flooding may increase downstream of
channelization works: flood waves attenuate less because
flows are delivered downstream more efficiently; excess
water is no longer stored temporarily on upstream floodplains;
and the contrast in conveyance at the end of the channelized
reach allows flow to back up in this area. Brookes (1987)
described channel enlargement in British rivers in response
to increased peak discharges from channelization.
Other topographic modifications may divert tributaries
from one stream to another. Road crossings and mine tailings
can dam and reroute streams, hydroelectric developments
may divert runoff to increase head, and irrigation ditches
may return flow to other channels. In these cases, flow is
reduced in the original channel while the new channel must
accommodate higher flows. More minor diversions may
result from rerouting of road drainage or formation of ruts
on ORV tracks, cattle paths, and hiking trails.
Impoundments may decrease water yield by increasing
the proportion of runoff lost to evaporation, and they strongly
affect downstream hydrographs even if they freely pass
flows. Impoundments usually retard and decrease the
amplitude of flood waves because they provide some storage
for flood waters and because outflow rates do not respond
instantly to changes in inflow. The frequency of high,
sediment-flushing flows is usually reduced in downstream
channels. A variety of other effects depend on the purpose of
the reservoir. Where flows are controlled for power generation,
diurnal flow variation may be extreme as power demands
fluctuate. Reservoirs built to supply water for consumptive
use usually decrease downstream water yields. At the
beginning of the runoff season, peak flows are reduced
downstream of these impoundments until storage capacity is
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
reached, but later peaks are allowed to pass. Flood control
reservoirs, on the other hand, are intentionally drawn down
before flood season, and flood peaks are temporarily stored
and released as lower flows. Most reservoirs serve multiple
purposes, and management strategy is modified to
accommodate a variety of demands.
Impoundments that enhance downstream baseflows by
seepage or release can promote riparian vegetation and allow
banks to stabilize and encroach. In a case reported by Bergman
and Sullivan (1963), the result was decreased channel capacity
downstream and a corresponding increase in flood severity. Yost
and Naney (1975) reviewed the effects of dam seepage on channel
morphology, water tables, salinization, and other characteristics.
Woody debris modifies channel flows by forming
obstructions and by triggering the formation of scour pools,
and this increased channel roughness slows and deepens
flows. Heede (1972) found that 70 percent of pools in a
Colorado forest stream were associated with organic debris,
and MacDonald and Keller (1987) measured a 2.5-fold
increase in flow velocity after natural organic debris was
removed from a forest stream. Some organic debris armors
banks and prevents channel widening, while other pieces
deflect flow toward erodible banks. Debris jams often force
channels to take new paths.
Altered transport of watershed products such as sediment,
organic debris, and chemicals also affects hydrology. Input
of fine sediments can decrease channel bed permeability,
and Bjerklie and LaPerriere (1985) and Bull and Scott (1974)
warned that sedimentation from in-stream mining can reduce
groundwater recharge. Coarser sediments also accumulate
in channels. Collier and others (1964) described infilling of
pools by sediment eroded from coal mine spoils, and Orme
and Bailey (1971) documented valley aggradation of up to
10 m at a site in southern California after erosion rates
increased upstream. These changes in channel morphology
affect water transport rates, and aggradation can increase the
frequency of overbank flooding. Hess (1984) attributed
increased flooding in Cabin Creek, Washington, to increased
input of sediment and organic debris from logging.
Rapid rates of channel aggradation can mantle stream
beds with permeable sediment, and low flows may trickle
through the bed sediment rather than flowing over it. Surface
flow may become discontinuous or ephemeral at these sites
even though discharge has not changed. Altered sediment
input rates also change the size distribution of sediments on
the channel bed (Dietrich and others 1989), and so can alter
flow velocity by modifying channel roughness.
Water Budgets and Runoff Modeling
Changes in basin hydrology are often assessed using water
budgets. A budget quantifies water inputs, changes in stor-
age, and export from a defined area by using the principle
that input, less change in storage, equals output. The effects
of particular land-use activities on hydrologic conditions
can be estimated if each component is defined and control-
ling processes are understood.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Water budgets can be constructed from field measurements
where sufficient data exist. Lewis and Burgy (1964), for
example, used measurements of precipitation, runoff, and
groundwater levels at a field site in the Sierra Nevada foothills
to back-calculate consumptive use of water by native
vegetation. Groundwater outflow removed about 5 percent
of the precipitation input to their study area. Available data
are rarely sufficient for a complete analysis, however, and
balances are usually calculated using models to quantify
particular components. Thornthwaite and Mather (1957)
outlined a method for calculating water balance, and the
method was further discussed and illustrated by Dunne and
Leopold (1978).
Precipitation is assessed using field measurements or
regional records. If nonlocal measurements are used, they
must be adjusted to reflect the characteristics of the study
watershed. Precipitation usually increases with elevation,
but rates of change are often difficult to evaluate because
high-elevation data are rare. Other inputs may include artificial
imports or contributions from regional groundwater.
Output is in the form of transpiration, evaporation, artificial
exports, basin outflow, and contributions to regional
groundwater. Transpiration and evaporation can be estimated
using methods described by Thornthwaite and Mather (1955)
and Penman (1948), and artificial exports are usually gauged.
Groundwater recharge generally presents the most intractable
unknown, and outflow is usually the component solved for.
Outflow can often be estimated using regional water yield
measurements.
Changes in storage are often small relative to input and
output, and include changes in soil moisture, changes in volume
of water in transport or in ponded storage, and changes in local
groundwater volume. Although seasonal variation of these values
is often high, year-to-year variation is low and is often ignored.
Mahacek-King and Shelton (1987) constructed a water
budget to estimate the effects of logging on peak discharges
in Redwood Creek. In this case, modelling consisted merely
of calculating Thornthwaite evapotranspiration and including
an estimated runoff delay factor calibrated using pre-logging
records. Post-logging flows were then reconstructed assuming
natural vegetation covers, and comparison of predicted and
measured flows showed an increase in peak flows.
Models for predicting hydrologic change as a function of
watershed conditions are increasingly common. Ng and
Marsalek (1989) used a hydrologic simulation program
developed by the EPA to assess likely effects of urban
development on water balance, and Simons and others (1981)
developed a runoff model specifically for use on forest roads
and calibrated it for conditions in California. Leavesley and
others (1987) described the Precipitation Runoff Modeling
System (PRMS), a general runoff simulation model currently
under development by the U.S. Geological Survey. PRMS is
being calibrated to predict the effects of timber management
on water yield and peak flows in coastal Oregon and the
Oregon Cascades. El-Kadi (1989) reviewed existing watershed
models and noted that few adequately evaluate groundwater
components.
63

Effects of Environmental Change
on Sediment
The amount of sediment produced, its mode and timing of
production, its grain size, and its transport through a channel
network are important characteristics of a watershed’s sedi-
ment regime. Changes to any of these factors can alter
watershed function.
Many studies provide measurements of erosion from various
types of land use. Results can be attributed to particular
processes at specific sites if sources are carefully isolated, but
many studies measure the combined effects of several activities
by monitoring only a watershed’s outflow. These studies can
measure impacts in that basin and indicate what responses are
important, but the effects of specific changes can rarely be
identified and results cannot be generalized. Dunne (1984)
discussed methods of predicting erosion in forested areas,
Larson and Sidle (1980) described erosion measurements
from the Pacific Northwest, and Toy (1984) reviewed the
effects of strip and open-pit mines on sediment production.
Erosion and Sediment Transport on
Hillslopes
Erosion rates are controlled by properties of erosive agents
and by those of erodible materials, and both are influenced
by environmental parameters. Most erosion processes ac-
tively transport sediment, though the processes that initially
dislodge sediment are often not the same as those that even-
tually contribute it to streams. For example, a sand grain may
be eroded by rainsplash and carried downslope by sheetwash.
Plants shield erodible materials from erosion by wind,
rain, and running water, and they modify the erosivity of
those agents by decreasing air and water velocities. Plants
also impart root cohesion to soils and increase their resistance
to erosion. Erosion rates usually change when vegetation is
altered, and changes in soil properties also affect erosion.
Compaction generally reduces soil erodibility but increases
the erosivity of flows, and changes in soil texture and structure
affect erodibility. Topographical changes can alter slope
stability and flow erosivity by changing the gradient of a
surface or diverting flow to a new path.
Because water is an almost ubiquitous factor in erosion,
most hydrologic changes alter erosion rates. Each type of erosion
process reacts differently to changes in environmental parameters.
Surface Erosion
Surface erosion includes processes of rainsplash, sheetwash,
rilling, and dry ravel (the progressive shedding of particles
from materials that are losing cohesion). Extensive experi-
ments by Federal, State, and university researchers in the
1950s and 1960s produced the Universal Soil Loss Equation
(USLE), which calculates surface erosion by sheetwash and
rilling as a function of hillslope gradient, soil type, slope
64
length, rainfall regime, conservation practice, and vegeta-
tion cover (Wischmeier and Smith 1978). The equation dem-
onstrates that surface erosion rates increase with decreased
ground-cover vegetation and soil organic matter or increased
hillslope gradient and length. Although originally developed
for agricultural lands, the equation has been used on range-
land (Renard and Foster 1985, Savabi and Gifford 1987),
forest land (Dissmeyer and Foster 1981), construction sites
(Holberger and Truett 1976), and roads (Farmer and Fletcher
1977, Meyer and others 1975). The equation’s assumptions
and limitations must be understood if results are to be valid,
and Wischmeier (1976) provided guidelines for appropriate
use. Williams (1975) modified the USLE to calculate sheet
erosion for individual storms and to estimate the proportion
of sediment redeposited.
Researchers are currently developing a more process-
based equation to take the place of the USLE for estimating
sheetwash and rill erosion. Known as the Water Erosion
Prediction Project (WEPP), this effort is intended to provide
an equation for application to many land uses (Lane and
others 1988, Nearing and others 1989).
Sheetwash, rill, and gully erosion rates usually increase
where soil compaction or loss of a permeable horizon increases
the volume of overland flow. Mechanical disruption of the
soil surface can inhibit surface erosion by reducing overland
flow, but it can also accelerate erosion by producing erodible
sediments, and disruption of organic soil horizons can accelerate
soil loss by exposing more erodible subsurface soils. Soil
erodibility generally increases as organic matter decreases
(Wischmeier and Smith 1978), so any activity that depletes
surface soils of organic matter is likely to promote erosion.
Vehicular traffic and livestock trampling can compact
subsurface soil horizons while producing a surficial dust
layer. The resulting sediment loss can be particularly high
because these changes increase both overland flow and
transportable sediment. Disaggregation can also promote wind
erosion, and Campbell (1972) suggested that roads in rural
subdivisions could be important sources of wind-blown dust.
Construction of long or steep slopes can increase rates of
sheetwash erosion and rilling, as is common on roadcuts.
More subtle topographic changes, such as furrows produced
by noncontour plowing or hiking trails, can also channel
enough runoff to trigger rilling. Dry ravel is common on
excavated soil faces, and Megahan (1980) measured the
combined effects of dry ravel and other surface processes on
roadcuts in Idaho.
Changes in the chemical environment generally have only
minor effects on the sediment regime of a watershed. Chemical
changes can make a soil more or less erodible by changing
its propensity toward aggregation. Some clay-rich soils, for
example, form clods that disintegrate rapidly when wetted,
while others produce clods that can be transported for tens of
meters as bedload in streams. This behavior is in part
determined by chemical properties of fluids in soil pores, and
Heede (1971) found the greatest incidence of dispersion-
related piping at a site in Colorado in soils with high
exchangeable sodium contents.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 The process of sheetwash erosion has been widely studied,
and erosion rates or trends can now be predicted for a variety
of conditions and for several types of hydrologic change,
although some uncertainties persist. Abrahams and others
(1988) and Govers (1987) both defined conditions under
which erosion was initiated, but with conflicting results.
Bryan (1979) examined the influence of gradient on sheetwash
and rainsplash erosion rates and found that antecedent moisture
and soil type influenced the form of the relation and that
unexplained variability was high. Govers and Rauws (1986)
evaluated the effects of surface roughness on transport, and
Lyle and Smerdon (1965) studied the influence of soil
properties on erosion rates. Kilinc and Richardson (1973)
used a rainfall simulator to examine the mechanics of
sheetwash erosion and developed a transport equation. Moore
and Burch (1986) found that sheetflow transport capacity
was well predicted by the product of flow velocity and
gradient (unit stream power). Most of these studies concerned
rain-generated flow and took into account the combined
effects of sheetwash and rainsplash. Hart and Loomis (1982),
however, examined surface erosion during snowmelt and
discussed problems of applying the USLE to snowmelt runoff.
Surface erosion selectively mobilizes fine sediment, so
delivered sediment may not reflect the size distribution present
in the eroding soil. Gilley and others (1987) found that the
grain size distribution of sediment in transport varied along a
slope profile, and Lu and others (1988) examined selective
transport and deposition of different sediment size classes.
Duncan and others (1987) measured the size classes of road-
surface sediment delivered to streams through ephemeral
channels at a site in western Washington and found that sands
tended to accumulate in the channels while clays were washed
on through. Hamlett and others (1987) demonstrated that
during large storms, differences in character decrease between
sediment available for transport and that removed by erosion.
Persistence of accelerated surface erosion often depends
on how quickly vegetation recovers. Megahan (1974) found
that erosion rates on roadcuts and fill slopes in the Idaho
Batholith decreased exponentially with time as sites recovered.
Gullies
The distinction between rills and gullies is based on the size
of the feature: incised channels that cannot be plowed over
are traditionally considered gullies while smaller ones are
called rills. Gully development represents several erosion
processes. Gully widening is accomplished by sheetwash
and ravel on walls, in-channel tractive erosion, and small-
scale landsliding, and incision is a form of channel erosion.
Melton (1965) described the potential importance of
vegetation in controlling gully initiation and hypothesized
that arroyo formation in the American Southwest may have
been promoted by decreased swale vegetation. Heede (1971)
examined processes of gully development on two soil types
in Colorado and found that pipe collapse formed gullies
more readily on soil with high concentrations of exchangeable
sodium. Soil compaction is often a factor in gully development
because it promotes surface runoff.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Topographic modification by road construction or skid-
trail use is the most common cause for gullying in logged
areas. Gullies often form where drainage is diverted onto
unprotected slopes by roadside ditches and culverts, where
culverts block and divert flow over roadbeds and fill slopes,
or where ruts concentrate flow.
Most changes affecting gully erosion involve altered
hydrology. Either runoff is increased and increases erosion
power, or channel networks are modified or expanded and
expose susceptible sites to erosion.
Landslides
Mass-wasting processes attack the entire soil profile. Plant
roots inhibit shallow landsliding and soil creep by increasing
soil cohesion. Swanston (1970) calculated failure conditions
for a debris avalanche in the Maybeso Valley of southeast-
ern Alaska and evaluated importance of root cohesion, and
Swanston (1969) showed that root decay rates may explain
the delay observed by Bishop and Stevens (1964) in land-
slide initiation after clearcutting in the same area. Hawley
and Dymond (1988) found that failure frequency increased
with distance from trees at a site in New Zealand. Transpira-
tion by vegetation can reduce activity levels on deep-seated
failures by decreasing soil moisture.
Construction and mining activities modify topography
and often create cut or fill slopes that are steeper and more
unstable than those originally present. Sites where
oversteepened materials are newly emplaced or where drainage
has been disrupted are at particular risk, and these conditions
are common along roads and on tailings piles. Failure generally
occurs as a debris avalanche or slump in these cases. Hagans
and Weaver (1987) estimated that 40 percent of erosion in
the Redwood Creek watershed resulted from road-related
diversions and channel crossings. Construction can also
accelerate earthflows by loading the head of the flow or
undercutting the toe, and roadcuts in earthflow terrain often
gradually encroach on roads.
Undercutting of rock slopes occasionally destabilizes
bedrock and causes slides that range in scale from single
dislodged blocks to rock avalanches of several thousand
cubic meters. Mining activity on Turtle Mountain in Canada
may have helped trigger a slide that brought down 30 to 90
million metric tons of mountainside (Selby 1982).
Much research has been devoted to developing methods
of recognizing unstable sites and evaluating site stability.
Rice and Lewis (1986) and Rice (1985) evaluated landslide
risk by statistically analyzing the settings of slides in northwest
California, and Duncan and others (1987) used a similar
method in Oregon and Washington. Rollerson and others
(1986) measured landslide frequency as a function of terrain
class to develop a risk rating for sliding in coastal British
Columbia. Landslide risk varies with time after an activity
has been completed. If organic debris is incorporated in fills,
road-related slides may occur after it decays, and failures on
clearcuts may be delayed until root networks have rotted.
Ziemer and Swanston (1977) measured decreases in hemlock
and spruce root strength as a function of time after clearcutting.
65
The National Council of the Paper Industry for Air and
Stream Improvement (NCASI 1985) reviewed processes of
landsliding in forested sites, discussed its relation to timber
management, described methods of assessing landslide risk,
and summarized results from a variety of landslide inventories
in the Pacific Northwest.
Landslide scars are susceptible to surface erosion as long as
they remain bare, and landslide frequencies at a site depend in
part on how rapidly landslide scars heal. Several studies have
examined rates of site recovery after landsliding by describing
vegetation changes (Smith and others 1984) and changes in soil
character (Trustrum and DeRose 1988) through time.
Because pore-water pressure is so important in triggering
landslides, hydrologic changes that increase pore pressures
often cause slides. Swanston (1967) measured piezometric
head as a function of precipitation at several Alaskan sites
and found that head increases were higher in drainage
depressions, thus providing an explanation for high rates of
landsliding in swales. Caine (1980) compiled rainfall
intensities at 73 sites throughout the world as a function of
duration for storms that triggered debris flows and identified
threshold values for landslide initiation on undisturbed slopes.
Soil Creep
Soil-creep rates are likely to increase with devegetation and
regrowth on hillslopes. Tunnels left by decaying roots are
filled by particles falling vertically rather than perpendicu-
larly to the hillslope surface, and this displaces the particles
slightly downslope. In addition, creep may be promoted by
loss of root cohesion, and soil moisture increases from de-
creased transpiration are likely to prolong the season for
plastic creep. Barr and Swanston (1970) and Dedkov and
others (1978) found that rates tended to be higher at wetter
sites or during wetter periods. Accelerated creep may in-
crease soil accumulations along channel margins and in-
crease bank erosion rates.
Saunders and Young (1983) compiled creep measurements
and indicated the range of rates that might be expected, but
measurements are difficult to carry out and so are rare. Accurate
measurements require long monitoring periods, and most mea-
surement techniques either are not sensitive to vertical
discontinuities (e.g., inclinometer tubes) or can be used only
once because they require disruption of the profile to record a
measurement (e.g., columns of marked particles). Most
measurements of creep rates after a change in land use either have
been of too short a duration to provide significant results or are in
rapidly creeping earthflows rather than typical hillslope soils.
Treethrow
Uprooting of trees disturbs large volumes of soil, and rootwads
are usually displaced downslope. This contributes to
downslope soil transport and exposes soil to further erosion
by sheetwash and rainsplash, but transport rates by treethrow
have rarely been measured. Treethrow is particularly com-
mon in riparian zones, where trees tend to be short-lived,
roots are shallow, and banks are susceptible to undercutting
(McDade and others 1990).
66
Sediment Delivery to Streams
Much of the sediment eroded on hillslopes is redeposited
elsewhere on the slope. For example, Vice and others (1968)
estimated that over half the sediment mobilized during free-
way construction at a site in Virginia remained within the
watershed. Although surface erosion rates are widely mea-
sured, much less is known of sediment delivery rates to
channel networks. Williams (1975) incorporated a sediment
delivery term into the USLE, and Tollner and others (1976)
developed an equation for estimating deposition from sheetflow
as a function of flow character, vegetation character, and
transport distance through vegetation. Heede and others (1988)
examined the role of vegetation recovery after a chaparral
fire in controlling the timing and rates of sediment delivery
to streams, and thus in controlling the timing and location of
channel adjustments. Haupt (1959) identified factors that
control the redeposition of sediment downslope of logging
roads, and Cook and King (1983) measured the volume of
sediment filtered out by windrows on fill slopes below roads
at a site in Idaho. Khanbilvardi and Rogowski (1984) and
Novotny and Chesters (1989) reviewed methods of estimat-
ing delivery ratios on the scale of plots and hillslopes.
Sediment delivery in large watersheds has been correlated
with morphological factors. Roehl (1962), for example, found
that the proportion of sediment eroded on hillslopes that
arrives at a watershed’s mouth decreases with increasing
watershed area and channel length, and increases with
increasing relief ratio. This relation implies that some sediment
is lost in transport and may reflect lowland aggradation or
chemical dissolution during transport and storage.
Erosion, Transport, and Deposition
in Channels
Changes in a channel’s ability to erode and transport sedi-
ment alter the channel’s morphology, and altered morphol-
ogy affects transport. Although some mechanisms and inter-
actions between erosion, transport, deposition, and morpho-
logical change are well known for low-gradient alluvial
streams, their expression in the small, high-gradient chan-
nels characteristic of forested watersheds is poorly under-
stood (see Lisle 1987 for discussion).
The extent of major channel changes is usually measured
using sequential aerial photographs. Grant (1988) described
techniques for assessing channel changes using aerial
photographs and hypothesized relationships between the
pattern of a channel alteration and the type of process
generating it. Diagnostic features include the presence or
absence of landslide scars, the length and continuity of the
change, and whether the magnitude of the change increases
or decreases downstream.
Bank and Channel Erosion
Processes active on stream banks include surface erosion,
small-scale landsliding, and tractive erosion from channel
flows. Rates of bank erosion in alluvial soils have often been
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
measured (e.g., Hooke 1980), but measurements from bed-
rock or colluvial banks are less common. Hooke (1979)
described factors controlling rates of surface erosion,
landsliding, and tractive erosion on banks. Pizzuto and
Meckelnburg (1989) related bank erosion rates along
Brandywine Creek in Pennsylvania to near-bank flow ve-
locities and also noted the importance of bank vegetation.
Roots enhance bank stability by increasing soil cohesion;
ground cover shields erodible sediment from flow; and stems
and leaves slow flows and decrease their erosivity. Smith
(1976) found that removal of bank vegetation in the Alexandra
Valley, Alberta, increased erosion rates by a factor of 20,000.
Unstable banks allow channels to widen and become
shallower. Zimmerman and others (1967) showed that channel
morphology in a small stream that drains forest and grassland
varies with bank vegetation: forested reaches tend to be
wider than those with grass on their banks. Similarly,
Murgatroyd and Ternan (1983) found that streamside canopies
in a Dartmoor forest plantation were dense enough to inhibit
ground cover on banks, and channels widened in response.
Graf (1979) showed an association between the location of
incised channels and sites of low riparian biomass, though it
is unclear whether lack of vegetation promoted incision or
incision inhibited vegetation. Some Welsh streams have
widened because toxic chemicals from mining have stripped
vegetation from streambanks (Lewin and others 1977). Bray
(1987) measured vegetation-related changes in channel
morphology in a channelized reach in New Brunswick, where
excavation of a new channel without riparian vegetation
triggered rapid widening and formation of midstream bars.
As riparian communities redeveloped, the channel began to
reassume its original form. Kondolf and Curry (1986)
attributed widening of the Carmel River in part to decreased
bank vegetation, and Graf (1978) attributed narrowing of
Colorado Plateau rivers to increased bank vegetation.
Vegetation also affects channel morphology after it dies:
logs are important structural elements in forest streams.
Woody debris decreases flow velocity by roughening
channels; it forms obstacles that induce pool formation; and
it deflects flow into or away from banks. Fallen trees also
divert flow into new paths, and Keller and Swanson (1978)
found that meander cutoffs on moderate-sized, low-gradient
streams in the Midwest are often associated with fallen trees.
Logs no longer enter channels where riparian forests are
removed, and morphology must adjust to a decreasing
frequency of obstructions as the remaining logs decay. Hogan
(1987) associated simplification of channel morphology with
decreased input of organic debris after logging in the Queen
Charlotte Islands. Sedell and Froggatt (1984) described
extreme decreases in wetted margin along the Willamette
River that were due, in part, to removal of organic debris for
navigation. Organic debris had helped to maintain the
Willamette’s sloughs and backwaters, and many channels became
choked with sediment when logs and rootwads were removed.
Woody debris was cleared from channels to “improve”
habitat and decrease debris flow risk in many areas logged
10 to 30 years ago, and the resulting disruption of channel
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
morphology was extreme. Many of these channels now
resemble gravel flumes: they are without significant pools
and have mobile, unstable beds. Debris removal also
contributed to sediment loads by liberating the sediment
stored behind logs (Klein and others 1987).
Structures designed to armor banks or deflect flow are
often emplaced to protect floodplain developments from
channel migration. Deflected flows may impinge on a
downstream bank and accelerate erosion there. Direct
disturbance of stream beds by gravel mining, road and bridge
construction, trampling, ORV use, dredging, and other uses
mobilizes fine sediments and elevates turbidity. Burns (1970)
measured sedimentation during logging in northern California
and found that the highest rates accompanied debris removal
or use of heavy machinery in or near streams. Dredging or
gravel mining can provoke channel incision as erosion
compensates for loss of bedload (Collins and Dunne 1989).
Sediment previously deposited by a river and held in
storage in floodplains or terraces can be remobilized by
channel erosion. Kelsey and others (1987) used transitional
probability matrices to evaluate inputs from storage elements
with different characteristic stabilities in Redwood Creek.
Altered channel morphology has often been associated
with particular types of land use such as grazing (e.g.,
Kauffman and others 1983a) and dam construction (e.g.,
Petts 1979). Rango (1970) predicted the effects of cloud
seeding on channel morphology by examining channel
geometry as a function of rainfall at 673 sites along a climatic
gradient. Richards and Greenhalgh (1984) described problems
with evaluating nonequilibrium channels to understand the
impacts of land use, and warned that most channels are not
completely in equilibrium.
Flow leaving impoundments is depleted of sediment
relative to its carrying capacity and can therefore entrain
fine sediments. Pemberton (1976) used sediment transport
equations to predict the approximate volume of sediment
lost from the bed of the Colorado River after closure of Glen
Canyon Dam. Trapping of sediment in Flaming Gorge
Reservoir has caused erosion of fine sediments downstream,
but decreased peak flows have prevented removal of coarse
sediments introduced by tributaries, and riffles have grown
at confluences (Graf 1980). Where channels downstream of
dams have erodible beds, the sediment imbalance may trigger
entrenchment that can propagate up tributaries (Germanoski
and Ritter 1987). Moglen and McCuen (1988) suggested
that downstream scour is important even on the scale of
sediment detention basins.
Channel Sediment Transport
Sediment transport rates and the mode of sediment transport
depend on discharge. Bed material is not mobilized until a
particular discharge is attained, so changes in peak discharge
can strongly influence both the bedload transport rates and the
proportion of the total load that is conveyed as bedload. In most
cases, however, bedload comprises less than a third of the total
load, and few non-sand-bedded channels convey more than 15
percent of their load as bedload (Vanoni 1975, p. 348).
67
 Considerable research has been devoted to developing
methods of predicting transport rates as a function of flow
characteristics and topography. Equations developed by
Einstein, Meyer-Peter and Müller, and Yalin are most
common, and Gomez and Church (1989) and White and
others (1978) reviewed the use of these and other equations.
Transport predictions have been consistently poor for
mountain streams, so Smart (1984) developed an equation
for use in high-gradient channels. Nouh (1988) modified an
existing equation to account for transport by nonsteady flow
in ephemeral streams subject to flash floods. However,
transport equations have rarely been adequately validated in
field settings. Results vary widely in accuracy at particular
locations, and extrapolation of any method beyond the
conditions for which it was developed usually invalidates
the application.
Suspended sediment is widely measured and has been
monitored at U.S. Geological Survey gaging stations for
decades, but bedload measurements are less common. The
temporal and spatial variability of bedload transport and the
logistical problems of sampling moving rocks during floods
make direct measurement difficult and complicate the
interpretation of results. Bedload is often estimated by
measuring suspended load upstream and downstream of a
settling pond, measuring the volume of sediment deposited
in the pond, and using these data to calculate the proportion
of infill that represents bedload.
Direct hydrologic modifications by climate change or
interbasin transfers alter the frequency of sediment-transporting
flows. If peak flows or runoff increase, streams scour their
bed and banks more effectively, erosion and transport rates
increase, and channels tend to incise. Abbott (1976)
documented widening and entrenchment when water imports
increased flow in a channel in Colorado, and Dzurisin (1975)
demonstrated entrenchment, armoring, and knickpoint retreat
where flow was augmented by stream capture at a site in
Death Valley. Where the frequency or magnitude of sediment-
transporting flows is decreased, aggradation is likely. Gregory
and Park (1974) observed a reduction in channel capacity
through aggradation and bar stabilization after a British
reservoir decreased flood peaks. Jackson and Van Haveren
(1987) used measured regional relations between discharge
and channel geometry to predict the effects of decreased
flows on channel morphology in Beaver Creek, Alaska.
Urbanization commonly increases runoff and peak flows,
and Neller (1988) measured bank erosion rates in an Australian
urban area to be 3.6 times higher than in a similar rural
stream, and rates of knickpoint retreat to be 2.4 times greater.
Graf (1975) found that streams draining a developing suburb
near Denver initially aggraded their floodplains, but then
entrenched when postconstruction sediment loads decreased.
Wolman and Schick (1967) described channel widening to
accommodate bar deposition from increased sediment input
during urban construction in Maryland. Nanson and Young
(1981) attributed a two- to three-fold increase in channel
cross-sectional area in urban areas near the Illawarra Escarpment
of Australia to the combined effects of increased peak flows
68
from urban runoff and decreased area of floodplain storage
from levee construction.
Sediment Deposition in Channels
Many land-use impacts involve deposition in channels and
resulting changes in channel morphology. Rates and loca-
tions of channel erosion and deposition are controlled by
spatial and temporal variations in both a channel’s ability to
transport sediment and the contribution of sediment from
hillslopes. High rates of sediment input from hillslopes can
cause localized aggradation in channels. Short (1987) docu-
mented extreme widening and aggradation after landslide-
triggering storms at Cuneo Creek, California, and Anthony
and Harvey (1987) described a channel’s change from me-
andering to braided after an extreme increase in bedload
transport in Colorado. Perkins (1989) measured erosion rates
of landslide deposits in western Washington streams and
found that 20 to 80 percent of the volume was removed
within 7 years. Hansen and Alexander (1976) introduced
sand into a Michigan channel and monitored the resulting
decreases in pool volumes, increases in channel width, and
decreases in channel roughness.
Long-term changes from chronic sediment inputs are also
important, but are harder to evaluate. Madej (1982) compared
geometries of disturbed and undisturbed channels in western
Washington to describe a change in form that reflects the
cumulative influence of a century of land use. Miller (1990)
recorded the distribution of contrasting channel responses to
sediment input and found that low-order Appalachian streams
tended to scour during floods while higher-order channels
aggraded or reworked deposits. Mosley (1981) found that
much deposition of coarse sediment in New Zealand streams
was localized behind woody debris, and that sediment transport
rates were strongly influenced by the collapse of debris dams.
Fine sediment is transported in suspension. Suspended
loads are often influenced more by sediment input rates than
by transport capacity. However, debris flows can overwhelm
channels with slurry-like concentrations of sediment, and a
portion of this sediment is likely be deposited near the input
site, at obstructions, and where channel gradient decreases.
Benda (1985) examined patterns of debris flow transport
along channels in the Oregon Coast Range and identified
slide locations likely to produce the most mobile flows.
Fine sediments can be deposited from suspension if flow
velocities decrease, as may occur on floodplains, in ponds or
reservoirs, or in gravel interstices of the channel bed. A
vegetative filter equation developed by Tollner and others
(1976) can be used to estimate deposition rates among plants.
Shapley and others (1965) found an exponential decrease in
suspended sediment with distance downstream of an input
point and inferred substantial deposition within the channel.
Duncan and others (1987) showed that over 55 percent of
fine sediment introduced into ephemeral channels during an
experiment in western Washington was trapped in the
channels, and Megahan and Nowlin (1976) found that 40
percent of hillslope sediment input from sandy Idaho batholith
soils was stored in channels.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Processes of fine sediment deposition among channel gravels
are poorly understood. Cooper (1965) noted that intergravel
deposition can occur even from high-velocity flows because
sediment-laden flow percolates into low-velocity pockets in
the gravel bed. Bed compositions are widely measured and
are occasionally correlated to land-use intensities or activities.
Cederholm and Reid (1987) and Adams and Beschta (1980)
related percent intergravel fines to extent of roads and logging
in watersheds. Platts and others (1989) documented changes
in percent fines associated with increasing and decreasing
sediment inputs from timber management over a 35-year
period in Idaho. Beschta and Jackson (1979) examined intrusion
of fine sediment in gravel-bedded laboratory flumes, and
Shapley and others (1965) described field observations after
introduction of sediment into a stream in southeast Alaska.
Lisle (1989) measured intergravel sedimentation as a function
of sediment transport rates in northwestern California channels.
Work on intergravel deposition was reviewed by Jobson and
Carey (1989) to identify physical factors that influence
sedimentation. Sedimentation was found to be most profound
in less active parts of the stream bed, and location of the
deposits within the bed was influenced by the grain-size
distribution of bed sediments.
Impoundments usually form efficient sediment traps. Coarse
sediment is deposited in a delta as it encounters ponded
water, and only suspendible sediments are carried on. Brune
(1953) described a method for calculating the proportion of
sediment trapped by a reservoir; Moglen and McCuen (1988)
evaluated the ability of small impoundments to trap sediment;
and Mahmood (1987) reviewed the extent and significance
of reservoir sedimentation. Formation of a delta at the head
of an impoundment decreases the gradient upstream and
decelerates flow, causing some upstream aggradation.
Dams affect deposition downstream by changing the
distribution of channel-modifying flows. Decreased peaks
from flood-control dams contribute to aggradation downstream
of tributary sediment inputs, and this effect can propagate
upstream (King 1961). Petts (1984) described aggradation
due to the cumulative effects of damming and drainage ditch
excavation in a forest plantation in England. A dam on one
tributary decreased sediment transport capacity at the
confluence just as excavation on the other fork increased
sediment input, and the confluence aggraded rapidly. Petts
(1979) and Williams and Wolman (1984) reviewed the effects
of dams on downstream channels.
Land Use and Sediment Yield
Many studies describe the effects of particular types of land-
use activities on sediment yields without differentiating be-
tween sediment sources. Sediment yields from logging and
roads are widely documented (table 12), and studies gener-
ally show a 2- to 50-fold increase over background levels,
with most of the increase associated with roads. Increases in
sediment input can be much larger at sites where landsliding
is particularly important. Sediment yields decrease rapidly
after road use is discontinued and logged areas regenerate,
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
so yields measured more than 5 years after logging are
usually less than five times higher than background rates.
Urbanization also provokes increased sediment yields.
Walling and Gregory (1970) reported 2- to 100-fold increases
in suspended sediment concentration below construction sites,
and Wolman and Schick (1967) estimated sediment yields of
700-1800 t per 1000-person population increase due to
construction near Washington, D.C.
Although sediment yields are an important indicator of
land-use impact, they are difficult to interpret without
information on how the sediment is produced. Van Sickle
(1981) compiled long-term annual sediment yields from
many Oregon watersheds and showed that year-to-year
variation in yield was large. Basin sediment yields must
therefore be measured for a long period if they are to accurately
represent a long-term average. Roels (1985) noted that many
local measurements and results of plot-sized experiments
cannot be used to estimate rates over larger areas because of
defects in the study design, and emphasized that all
measurement programs should be carried out within the
framework of a statistically valid sampling plan.
Sediment Budgeting and Modeling
In 1917, Gilbert introduced a new approach to evaluating
land-use impacts. He inventoried sources of sediment in
watersheds affected by hydraulic mining in the Sierra Ne-
vada foothills and estimated sediment production and deliv-
ery from each source. Gilbert used this information to esti-
mate total sediment input to the Sacramento River system,
and from this he could predict trends in sediment transport
and evaluate impact persistence. The approach was largely
forgotten until 1966, when Leopold and others used a simi-
lar method to evaluate sediment inputs in a New Mexico
watershed. Since then, sediment budgets have been con-
structed for many other sites (table 13).
A sediment budget is a quantitative account of sediment
input, transport, and deposition in a watershed (Dietrich and
others 1982). Component processes are identified, and process
rates are usually evaluated independently of one another. If
the effects of particular land-use activities on each process
are known, the overall influence of a suite of existing or
planned land-use activities can be estimated. Sediment
budgeting is particularly effective for evaluating
nonequilibrium situations, where channel loads do not
necessarily represent hillslope erosion rates. Trimble (1977)
used a sediment budget to show that modern sediment yields
in the southeastern United States reflect recent remobilization
of deposits left by long-abandoned erosive land-use practices.
Church and Slaymaker (1989) used a similar approach to
demonstrate the influence of Pleistocene landforms on present-
day measurements in British Columbia. In both of these cases,
sediment yield measurements alone would not have provided
enough information to identify the influence of land use.
Storage is often the most difficult component of a sediment
budget to define. Madej (1987) evaluated sediment storage in
Redwood Creek and used results to predict long-term residual
69
 Table 12—Studies of sediment production from roads and timber management. Symbols are explained below.

Treatment Measurements
Location Approach Scale F R T C L O S Y

Amaranthus and others 1985 OR A P R T L

Anderson 1954 OR SA M R T Y
Anderson and Potts 1987 MT T C R T Y
Aubertin and Patric 1974 WV T B T Y
Beasley 1979 MS T B T Y
Beasley and Granillo 1988 AR T M T Y
Beschta 1978 OR T B R T L
Beschta 1990 - O - F S Y
Bethlahmy 1967 ID E P T S
Bethlahmy and Kidd 1966 ID T P R S
Bilby and others 1989 WA T P R S
Bishop and Stevens 1964 AK D P T L
Biswell and Schultz 1957 CA D P F R S
Blackburn and others 1986 TX T M T S Y
Brown and Krygier 1971 OR T B R T L Y
Burroughs and King 1989 - O - R S
Burroughs and others 1984 ID E P R S
Carr and Ballard 1980 BC T P R S
Cook and King 1983 ID T P R S
Day and Megahan 1975 ID D P R T L
DeByle and Packer 1972 MT T P F T S
Diseker and Richardson 1962 GA T P R S
Duck 1985 SL D C R Y
Duncan and others 1987b WA,OR S P R L
Durgin 1985 L E P F S
Dyrness 1967 OR D P R T L
Dyrness 1970 OR T P R S
Dyrness 1975 OR T P R S
Fowler and others 1988 WA T M R T Y
Fredriksen 1970 OR T B R T L Y
Froehlich 1991 Po T P R S
Furbish and Rice 1983 CA S P R T L
Gonsior and Gardner 1971 ID S P R L
Grant and Wolff 1991 OR T B R T L Y
Gray 1970 - O - T L
Gray and Megahan 1981 ID TMS P F R T L S
Gresswell and others 1979 OR D P R T L Y
Hagans and Weaver 1987 CA D P R T C L S
Harden and others 1978 CA A P R T L
Harr and Fredriksen 1988 OR T B F R T Y
Haupt 1959 ID S P R S
Hawley and Dymond 1988 NZ D P T L
Heede 1984 AZ T B T S Y
Heede and King 1990 AZ T P T S
Helvey and Fowler 1979 OR T P T S
Helvey and others 1985 WA T B F L Y
Hess 1984 WA D C T C L
Hornbeck and Reinhart 1964 WV T P R S
Johnson and Beschta 1980 OR S P R T S
Johnson and Smith 1983 ID T P R S
Johnson 1988 SL T C R T Y
Kidd 1963 ID T P R T S
Klock and Helvey 1976 WA T B F L Y
Kochenderfer and Helvey 1987 WV T P R S
Krammes and Burns 1973 CA T B R L S Y
LaHusen 1984 CA S P R T L
Leaf 1974 CO T P R S
Lewis and Rice 1989 CA S P R T L
Lewis and Rice 1990 CA S P R T L
McCashion and Rice 1983 CA S P R L
McClurkin and others 1985 TN T B T Y
McNabb and Swanson 1990 - O - F L S Y
Megahan 1974 ID T P R S
Megahan 1975 ID T M R T L S Y
Megahan 1978 ID T P R S
Megahan and others 1983 ID A P R S

70 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

 Table 12—(Continued)

Treatment Measurements
Location Approach Scale F R T C L O S Y

Megahan and others 1986 ID T P R S Y

Megahan and others 1991 ID T P R S
Megahan and Bohn 1989 ID T P R T L
Megahan and Kidd 1972a ID T P R T S
Megahan and Kidd 1972b ID T M R L S Y
Mersereau and Dyrness 1972 OR T P F T S
Meyer and others 1975 WI MT P R S
Mosley 1980 NZ D P R L S
NCASI 1985 - O - R T L
Nolan and Janda 1981 CA T M R T Y
O’Loughlin 1972 BC S P R T L
Patric 1976 - O - R T Y
Reid and Dunne 1984 WA T P R S
Reid and others 1981 WA AT P R L S
Rice 1985 CA S P T L
Rice and Datzman 1981 CA S P R T C L
Rice and Gradek 1984 CA S P R T C L S
Rice and Lewis 1986 CA S P R T L
Rice and McCashion 1985 CA S P R L
Rice and Pillsbury 1982 CA D P T L
Rice and Wallis 1962 CA T B T Y
Rice and others 1979 CA T M R T C L S Y
Rice and others 1985 OR S P T L
Riekerk 1985 FL T B T Y
Roberts and Church 1986 BC DA P T C L O S Y
Ryan and Grant 1991 OR D M R T C L
Sartz 1953 OR T P F S
Schroeder and Brown 1984 OR S P R T L
Sullivan 1985 OR T C R T L Y
Swanson and Dyrness 1975 OR S P R T L
Swanston 1969 AK S P R T L
Swanston and Marion 1991 AK AS P T L
Swanston and Swanson 1976 - O - R T L
Swift 1984a NC T P R S
Swift 1984b NC T P R S
Trimble and Weitzman 1953 WV T P R S
Vice and others 1968 VA T C R S Y
Watts and others 1986 ID E P R S
Wolfe and Williams 1986 CA A P R T L
Wood and others 1989 LA E P T S
Wu and Swanston 1980 AK M P T L
Ziemer and Swanston 1977 AK A P T L
Ziemer and others 1991 CA M P R T C L

Location:
AK Alaska FL Florida NC North Carolina TX Texas
AR Arkansas GA Georgia NZ New Zealand VA Virginia
AZ Arizona ID Idaho OR Oregon WA Washington
BC British Columbia L Laboratory Po Poland WI Wisconsin
CA California MS Mississippi SL Scotland WV West Virginia
CO Colorado MT Montana TN Tennessee
Approach or method: Scale:
A Reconstructs past using air photos or B Paired basins or sites
records, or by sampling age distribution C Case study
D Based on descriptive measurements M Multiple basins or sites
E Experimental P Process study
M Modeling or calculated results
O General review of topic Treatment:
S Survey of multiple sites F Fire
T Long-term temporal monitoring study R Roads
T Timber management
Measurements:
C Channel-related erosion and gullying
L Landsliding
O Other erosion processes
S Surface erosion
Y Sediment yield

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 71

 Table 13—Sediment budgeting studies. Symbols are explained below.

Site Land use Cover Process Type

Reference ADGILMRU CFGRVW ABCDEFGILPQRSTW OHRSW

Caine and Swanson 1989 CO ...I.... ..G... ..CD......QRS.. .H.SW

Caine and Swanson 1989 OR ...I.... .F.... ..CD.F....QRST. .H.SW
Collins and Dunne 1989 WA .D...... ...R.. A...E..I..Q.S.. ..RS.
Dietrich and Dunne 1978 OR ...I.... .F.... A.CDEF....Q.S.. ...SW
Dietrich and others 1982 - ........ ...... ..C.....L..R... OH.S.
Duijsings 1987 Lu ...I.... .F.... .BCD....L.QRS.. .H..W
Gilbert 1917 CA ADGI.MR. CFG..W A.....GI.P.RS.. .H.SW
Kelsey 1980 CA ..G.L.R. .FG..W AB...F.IL...S.. .H.SW
Kelsey and others 1987 CA ........ ...R.. A......I..Q.... ..RS.
Lehre 1982 CA ...I.... .FG... ABCD..GIL.QRS.. .H.SW
Lehre and others 1983 WA ...I.... ....V. AB....GI..QRS.. .H.SW
Leopold and others 1966 NM ..G..... .....W ABC...GI.P.R... .H.SW
Madej 1982 WA ...IL.RU .F.... A.C.E.GIL.QRST. .H.SW
Madej 1987 CA ........ ...R.. A......I....... ..RS.
McLean and Tassone 1991 BC .D...... ...R.. A......I..Q.S.. ..R..
Megahan and others 1986 ID ......R. ...... A......I.P..S.. ...SW
Parker 1988 NE ........ ...R.. ..........Q.S.. ..R..
Phillips 1990 NC A.G.L.RU .FG... .B....G..P.RS.. .H.SW
Prestegaard 1988 PA A.G..... .FG... .B....G.....S.. .H...
Reid 1990 Tz A.GI..R. C.G..W .B....G....R..W .H...
Reid and others 1981 WA ......R. .F.... .....FG.L..R... .H...
Roberts and Church 1986 BC ...IL.R. .F.... ABC..F..L..R.T. .H.SW
Sing 1986 CA ........ ...R.. AB........Q.S.. ..RS.
Smith and Swanson 1987 WA ...I.... ....V. A.......L..R... .H.SW
Stott and others 1986 Sl ..GI.... .FG... .B........Q.S.. ..R..
Sutherland and Bryan 1991 Ke ..G..... .....W ABCD...I..QRS.. .H.SW
Swanson and others 1982 OR ...I.... .F.... ..CD.F..L.QRST. .H..W
Trimble 1983 WI A.G..... .FG... AB....GI.P.RS.. .H.SW

Location: Land use: Processes evaluated: Budget type:

BC British Columbia A Agriculture A Channel aggradation O Overview
CA California D Gravel dredging B Bank erosion H Hillslopes
CO Colorado G Grazing C Creep R River reach
ID Idaho I Inactive D Dissolution S Storage
Ke Kenya L Logging E Transport abrasion W Watershed
Lu Luxembourg M Mining F Debris flows
NC North Carolina R Roads G Gully erosion
NE Nebraska U Urbanized I Channel incision
NM New Mexico L Landslide
OR Oregon Cover type: P Pond aggradation
PA Pennsylvania C Chaparral Q Bedload
Sl Scotland F Forest R Surface erosion
Tz Tanzania G Grassland S Suspended load
WA Washington R River T Treethrow
WI Wisconsin V Volcanic blast W Wind
W Woodland

effects of floods and land use. Swanson and Fredriksen (1982)
demonstrated the application of sediment budgeting to evaluate
impacts of timber management, and Reid (1989) used sediment
budgets to predict the response of channel morphology in a
small California grassland watershed to altered stocking rates,
suburban development, and climatic change.
Models that predict sediment export from watersheds are
in increasing demand. Most existing models couple a USLE-
based sheetwash erosion equation with a flow-routing model,
and cannot be applied to sites with other erosion processes.
Those that incorporate sediment budgeting, however, are
more flexible. Watson and others (1986) constructed a model
for channel erosion in north central Mississippi by coupling
bank stability, sediment routing, and flow routing models. A
procedure developed by the USDA Forest Service (Cline and
others 1981) combines information from sediment budgeting,
plot studies, and long-term sediment yields in the Idaho
Batholith to characterize average erosion rates for timber
management activities in different land classes, and can be
used to estimate erosion rates for different planning options.
Effects of Environmental Change
on Organic Material
Land-use activities that affect input, transport, character, or
decay rate of organic material alter erosion, nutrient cycling,
and hydrologic processes both on hillslopes and in streams.
Organic debris on hillslopes alters rates of surface erosion,
provides a moisture reservoir, contributes nutrients, and sup-

72 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.

ports microbial action important in nutrient cycling. In chan-
nels, woody debris forms obstructions that at some sites
have been shown to slow flow by 60 percent (MacDonald
and Keller 1987), trap 15 years’ yield of sediment (Megahan
1979), provide 60 percent of total elevation drop along
streams (Keller and Tally 1979), localize scour to form
pools, and deflect flow into or away from banks. Leaves and
other small organic pieces supply nutrients to stream ecosys-
tems. Maser and others (1988) reviewed the role of woody
debris on forest floors, in streams, and along coasts.
Production of Organic Material from
Streamsides
Vegetation change alters the production of organic material,
and changes in riparian vegetation are particularly effective
in modifying organic inputs to streams. Logging can de-
crease input of stable woody debris to streams by removing
trees that would eventually become large snags under natu-
ral conditions. The largest, most geomorphically effective
boles are often removed even from streamside management
zones, and when the smaller hardwoods left to provide shade
eventually enter channels, they are quickly removed by de-
cay or carried off by high flows. McDade and others (1990)
measured distance travelled by incoming woody debris along
39 streams in western Washington and Oregon and used
results to predict future debris loading as a function of buffer
strip width, and Van Sickle and Gregory (1990) constructed
a model to predict inputs of woody debris on the basis of
measurements of tree frequency, size, and mortality in ripar-
ian zones. Windthrow often provides a burst of woody de-
bris from buffer strips soon after logging, but this depletes
the source of future wood. Toews and Moore (1982) mea-
sured sizes of wood pieces in streams before and after log-
ging at Carnation Creek on Vancouver Island and found that
average size decreased within two years by removal of stable
debris, breakup of large pieces, and introduction of logging
slash. Bryant (1980) found that an initial increase in debris
loading after logging in Alaska was followed by a decrease
as the logging debris destabilized natural debris accumula-
tions and washed them out. Swanson and Lienkaemper (1978)
discussed the effects of timber management on debris inputs
and suggested management strategies to lessen impacts. Log-
ging can also increase input of hardwood leaf litter by allow-
ing more light to reach riparian hardwoods.
Grazing is particularly intense in riparian corridors because
of their proximity to water and their support of green vegetation
during dry periods. Grazing and trampling can prevent
regeneration of riparian trees, and heavy use can disrupt
banks, force channels to widen, and trigger removal of riparian
stands by undercutting (Kauffman and Krueger 1984).
Riparian vegetation is occasionally removed to increase
water yield or to reduce flood hazard by improving channel
conveyance. Riparian vegetation is often sparse in developed
areas. Urban input of organic materials is usually through
storm drains and sewers, and contribution of woody debris is
rare. Agricultural areas also rarely preserve riparian trees, and
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
crop plants provide only fine organic materials. Roads in steep
terrain are often built on floodplains or along bedrock-walled
channels, where construction may require replacement of natural
channel banks by unvegetated riprap or grouted rock.
Gravel mining can greatly reduce inputs of stable organic
debris by widening rivers and suppressing bank and bar
vegetation. Widened channels may begin to braid, and woody
plants rarely survive long on braided river bars because
channels rework the deposits frequently. Where vegetation
is allowed to regrow, natural stands of mixed hardwoods and
conifers may be replaced by disturbance-tolerant species
such as alder. Inputs of large, stable boles then decrease
while deciduous leaf inputs increase.
Soil alterations can affect organic matter input if they
influence vegetation. For example, compaction can reduce
growth rates and decrease the size of in-falling trees, and
soil disruption can prevent regrowth of trees.
Topographical changes alter debris inputs by changing
the accessibility of waterways. Roads built on benches along
river canyons create topographic breaks that prevent transport
of organic materials from hillslopes to channels, and
urbanization replaces riparian habitats with concrete-lined
channels or subterranean conduits. Agricultural stream banks
are often leveed or protected by unvegetated riprap and
concrete linings to prevent channel migration, and
channelization for flood control or navigation has similar
effects. Flood-control works also weaken interactions between
channel and floodplain: the area capable of contributing
organic debris to the channel is reduced, and fewer trees are
uprooted by floods.
Vegetation can be killed and regrowth inhibited where
toxic chemicals accumulate. Eroded mining waste is often
redeposited on downstream floodplains and may suppress
riparian vegetation for decades (e.g., Lewin and others 1977).
Atmospheric pollutants can have similar effects. Mill effluents
have deforested wide areas around Sudbury, Ontario (Amiro
and Courtin 1981); acidification from industrial pollutants is
increasing tree mortality in many areas; and smog is injuring
vegetation in some parts of California (Omart and Williams
1979). In each case, an initial increase in input of organic
material to streams is expected as trees die, but future inputs
will decrease as woody plants are replaced by more tolerant
grasses and herbs.
Erosion processes often contribute organic debris to
channels. Bank erosion undermines riparian trees; landslides
deposit trees in channels along with their sediment loads; and
overland flow rafts fine organics into channels. Land use that
accelerates these processes can increase debris input to streams.
Hydrologic changes also affect organic matter inputs.
Channels may be left too dry to support riparian trees where
water is diverted, but riparian vegetation may be promoted
by reservoirs that reduce peak flows and heighten base flows
(e.g., Petts 1977). Seepage from dams and stockponds can
support riparian communities where seasonal desiccation
formerly prevented their survival (Yost and Naney 1975). In
general, changes that decrease diurnal or seasonal flow
fluctuation promote riparian vegetation, while those that
73
increase variability reduce vegetation by destabilizing
substrates and altering water tables.
Organic material can also be contributed by fauna. Insect
drift is an important food source for in-stream biota; beavers
import woody debris while building dams; and cattle
contribute manure. Some land-use activities intentionally
add woody debris to channels. Logs are placed to stabilize
banks, improve fish habitat, prevent channel incision, and
dissipate energy. In-channel structures such as docks, bridges,
and pilings may fail during floods and contribute wood.
Systemic changes such as acidification, air pollution, and
global warming can provoke replacement of sensitive species
by more tolerant ones, and inputs of dead wood are likely to
increase while plant communities are reacting to these changes.
In-Channel Production of Organic
Material
Much organic material originates within channel systems.
Fine organics are produced by in-situ biological activity and
are an important component of aquatic food chains, and
carcasses of anadromous fish once were important sources
of nutrients in streams and riparian areas of western forests.
Richey and others (1975) found that decomposition of re-
turning sockeye salmon in Tahoe basin streams still pro-
vides a significant nutrient subsidy to the channel ecosys-
tems. Management activities change in-situ production by
altering physical environments and nutrient inputs.
Activities that reduce riparian canopy cover promote in-
situ primary production by increasing water temperatures
and the amount of light reaching streams. Logging and grazing
can have this effect, as can floodplain development for
agriculture or urbanization. In each case, a long-term decrease
in woody debris input is accompanied by increased production
of fine organics. Where dense second-growth is reestablished,
productivity may drop below original levels as the young
canopy closes over the channel.
Input of nitrates, phosphates, and other nutrients associated
with urbanization and agriculture can increase aquatic primary
production. If increases are too high, biological activity depletes
dissolved oxygen and makes streams uninhabitable to respiring
organisms. Welch and others (1989) related increased algal
biomass downstream of urban areas in the Spokane River to
phosphorus inputs from urban sources, and Larsson and others
(1985) suggested that eutrophication of the Baltic Sea is
promoted by the cumulative contribution of nutrients from
municipal and industrial sewage delivered by rivers.
Inputs of fine sediment can reduce in-stream production
by coating substrates, thereby making surfaces unstable and
blocking light (Power 1990). Similarly, increased turbidity
reduces light penetration through the water column. Dredging
and other in-stream uses can also reduce substrate stability
and increase turbidity. Decay of silt-sized organic particles
in the channel substrate decreases survival of fish eggs,
alevins, and invertebrates by depleting oxygen from
intergravel environments. Sharpley (1986) measured high
nutrient concentrations associated with suspended sediment
74
and inferred that high sediment loads might accelerate
eutrophication.
Fluctuating flows downstream of water and power
developments usually restrict most in-stream production to
low-flow zones, with only those organisms tolerant of periodic
desiccation inhabiting channel margins. If reservoirs are
shallow and algae-rich, they may contribute to organic loads
downstream.
Modification of Organic Material
Many land-use activities modify decay rates of organic
material by altering vegetation communities, moisture dis-
tribution, and temperature. Edmonds (1980) measured litter
decomposition rates on forest floors in Washington, and
Harmon and others (1987) measured rates of wood decom-
position in the Sierra Nevada. Fogel and Cromack (1977)
showed that Douglas-fir litter in a western Oregon forest
decays more rapidly under wetter conditions, and Fenn and
Dunn (1989) found that litter decomposition rates increased
with ozone pollution in the San Bernardino Mountains.
Most research on decay has been carried out in terrestrial
environments, and little is known of decay rates for in-
stream logs. Organic material is also altered by being eaten
and metabolized or excreted.
In-Channel Transport of Organic
Material
How long a log remains in a channel depends on how well it
is anchored, how large it is relative to flows, and how long it
takes to decay. Different species provide wood of different
structural roles, and different parts of a channel system
interact differently with logs. In general, logs are progres-
sively less stable and of less structural importance with
increasing stream size. Swanson and Lienkaemper (1978)
found that logs were randomly distributed in first- and sec-
ond-order channels where streams are not large enough to
mobilize them, but most logs were transported into debris
jams in third- through fifth-order channels. Activities that
affect the size of peak flows or incoming woody debris can
alter transport of organic matter. Bilby and Ward (1989)
measured pools and sediment accumulations formed by logs
as functions of channel size and documented their decreas-
ing role as channel width increases, and Bisson and others
(1987) described debris mobility as a function of debris size
and channel width.
Some land-use activities suppress transport of woody
debris. Low-order channels along roads are often routed
through culverts that are too small to pass organic debris,
and debris may accumulate at the intakes. Not only does this
deprive downstream reaches of woody debris, but
accumulations can plug culverts and cause flows to overtop
and wash out road fills.
Where debris flows occur, large woody debris can be
mobilized even in low-order channels. Swanson and
Lienkaemper (1978) found 35 percent of the length of first-
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
order channels in a coastal Oregon basin to be scoured to
bedrock by debris flows. They warned that recovery of these
channels was unlikely because timber-managment-related
erosion rates remained high and logging had removed the
source of future large wood.
Some land-use activities remove woody debris once it is
in transport. Logs may be cleared from channels to aid
navigation, improve fish passage, reduce flood hazard, or
provide material for wood products. Impoundments decrease
inputs to downstream reaches by trapping woody debris.
Loss of woody debris can accelerate the transport of other
organic materials through the stream system. Cederholm
and Peterson (1985), for example, found that retention of
salmon carcasses in small channels on the Olympic Peninsula
was correlated with the frequency of woody debris.
Organic Material Budgets
There is growing interest in the construction of woody de-
bris and organic material budgets for watersheds, although
few have yet been measured. Hogan (1987) evaluated the
effects of logging on large organic debris by constructing
budgets for pre- and postlogging conditions, and Caine and
Swanson (1989) contrasted sediment and organic material
budgets for small watersheds in the Cascades and in the
Rocky Mountains. Cummins and others (1983) reviewed
components of organic-material budgets, described their con-
struction and philosophy, and compared budgets for five to
seven stream orders in four parts of North America.
Effects of Environmental Change
on Chemicals
Chemical inputs and transport are of concern primarily be-
cause of toxic effects and changes to nutrient cycles. Moody
(1990) provided a general review of the extent of groundwa-
ter contamination by chemicals in the United States; Smith
and others (1987) reviewed contamination of surface waters;
and Gilliom and others (1985) surveyed pesticide concentra-
tions in rivers. Nutrients are chemicals required by plants
and animals, and include nitrates, phosphates, and many
others. Although nutrients can be toxic and degrade water
quality in high concentrations, their major significance to
forest management is their role in plant growth.
Input and Removal of Chemicals
Chemical inputs from land use must be considered in the
context of the natural chemical balance. Many compounds
are naturally present in solution due to bedrock weathering.
Reynolds (1986), for example, measured solution load in a
catchment in Wales and found that 78 percent of the material
lost from the basin was dissolved.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Vegetation is a major component of nutrient and chemical
cycles, and many nutrients are stored in plant tissues until
released by decay. Long-term nutrient budgets in stable
ecosystems are balanced, and amounts returned to the
environment are sufficient to support regrowth. Edmonds
(1980) evaluated nutrient inputs from litter decomposition in
several forest habitats of western Washington, and Graham
and Cromack (1982) assessed nutrient content of decaying
wood in the Olympic rainforest. Little and Waddell (1987)
noted the importance of the forest floor and downed wood as
nutrient reservoirs in forests of the eastern Olympic Peninsula.
Waide and Swank (1976) described nutrient storage and
transfer through components of a plant community in the
southern Appalachians and illustrated the concept that
communities storing large amounts of nutrients tend to be
resistant to change under natural conditions, whereas those
that transfer nutrients rapidly between storage elements tend
to be resilient rather than resistant. Communities with stored
nutrients thus respond to isolated disasters with little change
in composition, while rapid-turnover communities may initially
change in character but quickly reassume their initial state.
Under natural conditions, sites depleted of nutrients by
catastrophes like fires are recolonized by species that require
low nutrient levels or can extract nutrients from sources
inaccessible to other plants. Less tolerant species return
when nutrient levels are reestablished. If land use repeatedly
devegetates a site and prevents regrowth of colonizing species,
the site may eventually be depleted of some nutrients. Forests
and croplands are particularly susceptible to loss of nitrogen
and phosphorus, and farmers and foresters avoid this problem
by adding fertilizers. Little and Waddell (1987) assessed the
chemical composition of undisturbed forests and soils and
tested seedling growth rates to ascertain that logging would
not cause nutrient deficiencies at a site on the Olympic
Peninsula.
Nitrogen can be depleted if large amounts of organic
carbon are introduced to soils. High carbon levels temporarily
lock up available nitrogen by promoting increased microbial
activity as materials decompose. Cochran (1968) used
measured nutrient concentrations and ratios to evaluate the
potential for carbon-induced nutrient deficiencies in a nutrient-
poor Oregon soil. In other cases, increased decay rates or
volumes of decaying material can increase nutrient availability
at a site. Burger and Pritchett (1988), for example, found
high nutrient levels where logging slash was chopped and
mixed into soils in Florida.
Many land-use activities introduce chemicals to watersheds,
but often their amount and location are poorly documented.
Markin (1982) found that less than half the insecticide applied
as an aerial spray to a forested area on the east slope of the
Washington Cascades landed on the intended plot, and Miller
and Bace (1980) found that measured herbicide exports in
streamflow corresponded to the estimated volume of aerially
applied herbicide pellets that fell directly onto surface water.
Brockway (1983) measured accumulation of nutrients and
heavy metals in a Michigan forest floor after sludge application,
and Harris and Urie (1983) measured these effects over a
75
longer period at another site in Michigan. Nitrate contamination
due to heavy fertilizer applications has polluted groundwater
in many agricultural areas. Staver and Brinsfield (1990)
examined seasonal changes in nitrate availability in corn fields
and used this information to suggest a fertilization schedule
that reduces contamination of groundwater by excess nutrients.
Other activities expose existing chemicals to the environment.
For example, LaPerriere and others (1985) found high levels of
heavy metals downstream of gold dredges. Mining also exposes
unweathered rock to leaching in tailings piles.
Transport of Chemicals
Chemicals can be transported as particles, in solution, or
adsorbed onto sediment and organic matter. Recent concern
over water quality and the disposal and leakage of hazardous
wastes has generated considerable research on chemical trans-
port in the environment. Much of this work is relevant to
chemicals used in forest management.
The properties of introduced chemicals must be known if
their impacts are to be understood. Chemicals that are insoluble
and nonreactive may accumulate where applied, while more
mobile species can contribute to off-site effects. Gilliom and
others (1985) noted that persistent, soluble pesticides are
likely to show up in river water, while persistent, hydrophobic
species are more common in riverbed sediments. Modes of
migration are also extremely important. Sophocleous and
others (1990) investigated movement of agricultural herbicides
and their breakdown during transport. Results demonstrated
the importance of cracks and permeability variations in
governing subsoil transport times and paths, and indicated
that diffusive models are insufficient for predicting transport.
Soil structure influences chemical leach rates by affecting
the infiltration and residence time of soil moisture, and Price
and Watters (1989) evaluated the role of soil horizons in
controlling chemical fluxes in an Ontario forest. Altered soil
structure and compaction also affect root penetration and
change the distribution of organic materials and nutrient
availability within soils. Organic compounds strongly influence
the mobility of chemicals, so land-use activities that alter the
organic content of soils modify their capacity to transport
chemicals and nutrients.
Because chemical transport is dependent on water
movement, any activity that alters runoff volumes or timing
also affects rates of chemical and nutrient transport. The
mode of runoff generation can also influence chemical
transport. Ford and Naiman (1989) examined the role of
groundwater in supplying dissolved nutrients to streams at a
forested site in Quebec, and Deverel and Gallanthine (1989)
associated the occurrence of saline groundwater with patterns
of irrigation, geomorphic features, and hydrologic conditions in
the San Joaquin Valley.
Many compounds are adsorbed by and transported with
silt or clay particles, so land-use activities and soil changes
that increase erosion also increase transport of chemicals
and nutrients. Alberts and Moldenhauer (1981) demonstrated
76
that cultivation methods that produce lower-density aggregates
lead to higher concentrations of nitrogen and phosphorus in
eroded sediment. Suspended sediment carries high
concentrations of bioactive phosphorus, a nutrient associated
with eutrophication of waterways (Sharpley 1986). Duffy
and others (1986) found that 40 percent of the nitrogen and
70 percent of the phosphorus exported from southern coastal-
plain watersheds was associated with sediment. Cerco (1989)
evaluated chemical and temperature factors that control adsorption
of nutrients on sediment particles.
Surface erosion preferentially removes the most nutrient-
rich part of the soil and leaves soils impoverished, and
erosion of chemically contaminated deposits can spread the
impact to wider areas. Roberts and others (1982) found that
the area of contamination rapidly increased as sediments
contaminated first by subsurface flows were dispersed by
sheetwash erosion.
Some chemicals have been evaluated for mobility in field
settings by applying the chemical and monitoring its
concentration in runoff. Davis and Ingebo (1973) found
picloram concentrations in runoff below a chaparral
conversion site that were high enough to damage downstream
crops, and they detected residues of the herbicide in soil 14
months after application. Michael and others (1989) and
Neary and others (1985) also monitored export of picloram
and found that detectable residues remained in soil solution
for at least 40 weeks at sites in the southeastern United
States. Similar studies have examined other commonly used
herbicides and pesticides. Miller and Bace (1980) measured
concentrations of forest herbicides in streams and found that
herbicide applied in pelletized form reached streams more
readily than that applied as a liquid spray. Pelletized chemicals
fell through the canopy to enter surface waters, while some
of the sprayed herbicide was trapped on foliage.
Monitoring studies indicate that logging usually causes a
short-term increase of nitrogen concentrations in soil solutions
and streams (e.g., Mann and others 1988, Sollins and
McCorison 1981, Tiedemann and others 1988). Increases
may result from reduced nutrient uptake, increased subsurface
flow, increased alteration of nitrogen compounds to leachable
forms, and increased volume of decaying organics. Vitousek
and Melillo (1979) reviewed these and other mechanisms of
nutrient loss. Baker and others (1989) measured release rates
of nine nutrients from logging slash during decomposition
and found that nitrogen was released most slowly. Other
activities that contribute to these mechanisms, such as
herbicide application (Feller 1989, Vitousek and Matson
1985) and urbanization (Gold and others 1990), may have
similar effects.
Urban land use contributes a variety of chemicals to
streams. Osborne and Wiley (1988) found stream nutrient
levels to be correlated to the proportion of a watershed
urbanized in Illinois, while Scott (1981) evaluated the
topographic, design, and management factors that affect
delivery of road salt to streams in Toronto. McBean and Al-
Nassri (1987) related distance travelled by road salt to average
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
vehicle velocity and stressed the importance of splash in
spreading the chemical impact. Lead from gasoline is also
transported in streams (Alexander and Smith 1988). Roberts
and others (1982) documented the migration of PCBs from a
highly contaminated spill site.
Streams in mining districts tend to have high chemical
loads, and Platts and others (1979) described the distribution
in an Idaho stream of toxic chemicals from an abandoned
mine. Most contaminants at this site emanated from mine
drainage and from leaching of spoils piles. Mok and others
(1988) and Mok and Wai (1989) measured concentrations in
groundwater and surface water, respectively, of arsenic species
near a mine in Idaho.
Many studies have explored the effects of land-use activities
on stream chemistry without examining how the contaminants
were transported to the streams. Tiedemann and others (1989)
found that grazing did not appreciably affect stream chemistry
at a site in eastern Oregon, and Johnson and others (1978)
found little chemical change due to grazing at a Colorado
site but noted a significant increase in bacterial counts.
Truhlar and Reed (1976) related pesticide concentrations in
streams to different types of land use in Pennsylvania, and
Mackay and Smith (1990) surveyed groundwater
contamination by agricultural pesticides in California.
Lawrence and Driscoll (1988) detected increased acidity
downstream of a New Hampshire clearcut and attributed the
change to increased soil nitrification. Increased aluminum
transport after logging was also noted during the study, and
concentrations lethal to fish were measured downstream of
the clearcut.
Alteration of Chemicals
Many chemicals are toxic when first introduced but break
down to harmless byproducts through biological activity or
weathering. The persistence of introduced species must there-
fore be understood if long-term or off-site effects are to be
predicted. Sophocleous and others (1990) examined the per-
sistence of atrazine, a crop herbicide, in soils and groundwa-
ter, and Sharom and others (1980) measured the persistence
of 12 pesticides in natural, distilled, and sterilized water to
assess the relative importance of biological and chemical
degradation. Norris and others (1977) measured breakdown
rates of the herbicide 2,4,5-T, and Neary and Michael (1989)
measured residues of a site preparation herbicide
(sulfometuron methyl) over time at a site in the southeastern
United States. Although data exist for particular chemical
species at particular sites, information has not yet been
generalized to the point that chemical persistence can be
predicted for the range of site conditions encountered.
Some practices alter existing chemicals and cause them
to interact with the environment in different ways. For
example, Helvey and Kochenderfer (1987) found that
limestone road gravels locally moderate the pH of runoff
from acid precipitation, and Birchall and others (1989)
demonstrated that aluminum released by acid precipitation
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
is less toxic to fish if streamwaters are enriched in silica.
Bolton and others (1990) found that vegetation conversion
alters the spatial distribution of nitrogen mineralization
because plants lend soils different mineralization potentials.
Introduced chemicals can modify the mobility and chemical
composition of chemical species already present. Acid rain,
for example, increases mobility of nutrients and may contribute
to nutrient deficiencies (Johnson and others 1982). Chemical
changes may also alter the ability of plants to make use of
available nutrients.
Burning increases nitrogen loss by transforming stored
nitrogen into more soluble compounds that are susceptible
to leaching (DeByle and Packer 1972, Grier 1975), and
increased leaching can raise nitrate levels downstream.
Burning also removes nutrients by volatilization, and Feller
(1988) related nutrient loss in British Columbia forests to
the proportion of forest floor consumed by fire.
Accumulation of Chemicals
Chemicals introduced to channels may be precipitated or
redeposited downstream. Many chemicals are adsorbed onto
fine sediment particles and accumulate where sediment is
deposited. Miles (1976) measured insecticide residues in
stream sediments in Ontario, and Gilliom and Clifton (1990)
detected high levels of discontinued pesticides, including
DDT, in sediments of the San Joaquin River. Imhoff and
others (1980) found that 31 percent of heavy metals intro-
duced from urban sources in the Ruhr watershed was re-
tained in stream sediments, and Marron (1989) showed ac-
cumulation of mining arsenic in floodplain deposits in Colo-
rado. Nutrients also accumulate at deposition sites. Johnston
and others (1984) calculated rates of nutrient trapping in a
Wisconsin wetland using measurements of sediment nutri-
ent contents and aggradation rates.
Aggrading impoundments can accumulate chemicals
adsorbed onto sediment, and reservoirs can also concentrate
dissolved pollutants if evaporation rates are high. Reservoirs
accepting irrigation drainage in California’s San Joaquin
Valley have become progressively enriched in selenium salts
and other toxic substances. Log jams also trap sediment, and
Bilby (1981) measured their effectiveness by monitoring
chemical fluxes in a New Hampshire stream before and after
removal of log jams. Triska and others (1989) injected nitrate
into a forested tributary of Redwood Creek and measured
outflow and changes in storage along the stream. Results
showed that 29 percent of the injected nitrate was held in
mineral storage and 19 percent in biotic uptake.
Repeated addition of a chemical to a site can cause a
cumulative increase in concentration. Lockery and others
(1983) measured particularly high levels of lead contamination
at dump sites for road snow, and Hofstra and Smith (1984)
found that salt concentration in soils decreased as a function
of distance from a road.
Evapotranspiration of chemical-rich irrigation water
precipitates salts in soil. Salinization is severe enough in the
77
San Joaquin and Imperial Valleys to prevent further
agricultural use of some tracts, and the effect is spreading.
Runoff from these areas adds to salt loads downstream, and
the impact is compounded where this water is used again for
irrigation. The Colorado River eventually becomes too saline
for further use because of heavy irrigation use.
Biota can accumulate and concentrate introduced
chemicals. Erdman and others (1978) and Neuman and
Munshower (1984) found high levels of molybdenum in
forage plants growing on coal mine dumps, and Nichols and
Scholz (1989) measured increased levels of radioactive
elements in trout downstream of a uranium mine in
northeastern Washington.
Nutrient Cycling, Chemical Budgets,
and Modeling
Chemical budgets can be constructed in much the same way
as sediment budgets. In this case, however, alterations to the
species present become an extremely important aspect of the
budget. Where the chemicals of concern are nutrients, entire
budgets can describe cycles of alterations at a single site as
nutrients are repeatedly incorporated into plant tissues and
released by decay. Mahendrappa and others (1986) reviewed
methods used to analyze nutrient cycling and described prob-
lems in budget construction. Hornbeck and others (1986)
constructed a nutrient budget to examine the effects of
clearcutting on a New Hampshire forest, and Gholz and
others (1985) assessed the influence of early stages of re-
growth on nutrient budgets after clearcutting in the H.J.
Andrews Experimental Forest of western Oregon.
Some budgets focus on particular chemicals. Waide and
Swank (1976) used nitrogen budgeting to predict forest
management impacts in the southeastern United States and
evaluated inaccuracies caused by lumping storage components
together. Mitchell and others (1989) studied the effects of
clearcutting on sulfur cycling at Hubbard Brook Experimental
Forest, New Hampshire. Ryan and others (1989) examined
inputs of sulfur to soils from atmospheric pollution in the
southeastern United States and found that soil adsorption rates
for sulfur are high but finite. They noted that rates of sulfur
loss to streams are already increasing and inferred that the
soil’s capacity is approaching saturation. Budget calculations
indicated that a steady state between atmospheric inputs and
runoff outputs would be reached in about 100 years.
Berndtsson (1990) constructed a chemical budget for sources
and sinks of pollutants downstream of Lund, Sweden, and
Imhoff and others (1980) used a similar approach to examine
heavy metal pollution in the Ruhr River. Larsson and others
(1985) constructed a nutrient budget for the Baltic Sea and
used it to demonstrate the cumulative importance of municipal
and industrial sewage delivered by rivers throughout the basin.
Researchers now often attempt to predict flow paths and
transport rates for chemical contaminants. Padilla and others
(1988) constructed a model to predict subsurface transport
of chemicals and found that transport is strongly affected by
the dependence of chemical breakdown rates on temperature.
78
The CREAMS model (Chemicals, Runoff, and Erosion in
Agricultural Management Systems; Knisel 1980) is one of
the most widely known methods for predicting chemical
transport from croplands. This model takes into account the
affinity of chemicals for sediment particles and also predicts
transport of sediment, but it does not calculate subsurface
transport and it requires calibration. Lorber and Mulkey
(1982) compared CREAMS with two other models for
predicting pesticides in runoff and found that all provided
valid estimates. Nutter and others (1984) found good
agreement between CREAMS predictions and measured
herbicide concentrations in forested basins for initial storms,
but the lack of a subsurface component caused underestimates
during later storms. Kenimer and others (1989) developed a
similar model that is claimed to have stronger hydrologic
and sediment transport components and can also take into
account the transfer of chemicals between adsorbed and
dissolved phases. Pennell and others (1990) compared the
predictions of five pesticide transport models to measured
distributions, and provided guidelines for selecting an
appropriate model. Emmerich and others (1989) compared
models for predicting transport by surface runoff and stressed
the importance of addressing nonhomogeneous conditions if
they are present.
Models have been constructed to predict chemical transport
rates in alluvial channels of lowland streams, but little work
has been done on mountain channels. Bencala and Walters
(1983) demonstrated that ordinary models for stream transport
do not work for pool-riffle sequences characteristic of steep
streams and developed a model that does apply to these
settings.
Chemical transport models must account for changes in
chemical state during transport. MacQuarrie and others (1990)
constructed a model that incorporates chemical and biological
breakdown of pollutants during groundwater transport, and
Tim and Mostaghimi (1989) and Padilla and others (1988)
modeled breakdown of pesticides during transport above the
water table. Jury and Gruber (1989) used a model to examine
the effects of soil variability on pesticide transport and found
that variability contributes to more rapid removal of pesticide
residues from the biologically and chemically active surface
soil layers, where most breakdown takes place. Results showed
that contamination by degradable chemicals may be more
widespread than expected because some toxic compounds
can persist for long periods once they leave the active zone.
Effects of Environmental Change
on Heat
Heat is transported through a watershed by changes in stream-
water temperature. Stream temperature is determined by the
temperature of inflowing water, radiant inputs and exports,
conductive inputs and exports, convective inputs and ex-
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
ports, and heat extraction from evaporation. Radiant inputs
are highest where infrared radiation impinges on the water
surface, and losses are highest where infrared radiation emitted
from the stream is not reflected back to the water. Conduc-
tive heat exchange is rapid where water is in contact with air
or substrate of contrasting temperature, and is accelerated
where the contrast is maintained by convective removal of
partially equilibrated air or water from the interface. Evapo-
rating water extracts heat from air and water to accomplish
the phase change. Beschta and others (1987) reviewed re-
search on stream temperatures and discussed changes caused
by timber management.
Vegetation influences all types of heat exchange in streams.
Tree canopies over streams decrease temperature fluctuations
by deflecting incoming infrared radiation and reflecting back
much of that emitted from the water surface. Because of
evapotranspiration and shading, daytime air temperatures
are generally lower under tree canopies than in clearings.
Differences between water and air temperature are thus less
under the canopy, and rates of conductive exchange during
warm parts of the day are reduced. There is also less convection
where temperature differences are small and where streams
are shielded from wind. Land-use activities that reduce riparian
cover usually increase daytime stream temperature and the
range of temperature fluctuation. Beschta and Taylor (1988)
and Holtby (1988) measured increased stream temperatures
associated with logging in the Oregon Cascades and on
Vancouver Island, respectively, although mechanisms
generating the change were not identified.
Altered soils can affect stream temperature both by causing
vegetation changes and by altering the relative importance
of overland and subsurface flow. Cold-season precipitation
is likely to be colder than groundwater contributions to
streams, but quickflow may be warmer than groundwater if
precipitation is warm or falls onto a warm surface.
Channel morphology influences rates of heat exchange in
streams by controlling the water surface area exposed to the
atmosphere. Channels often widen and aggrade to
accommodate increased erosion rates or altered peak flows,
and heat exchange is accelerated in wide, shallow channels.
Daytime temperatures and diurnal fluctuations are expected
to increase at such sites, especially if the widening has
removed riparian vegetation. Beschta and Taylor (1988)
found that stream temperatures remained elevated for several
years after major storms while riparian zones and bars
remained partially devegetated.
Whether a reservoir increases or decreases downstream
water temperature depends on the size of the impoundment
and the source of outflow. Broad reservoirs absorb more
infrared radiation than a channel and are less shaded by
vegetation, and downstream flows are warmer than usual if
outflow is removed from the reservoir surface. Cold water
collects at the bottom of deep lakes, however, so if outflow
is removed from the base of a dam, downstream flows are
colder than normal and have low organic nutrient contents.
Hubbs (1972) discussed the effects of dams on heat input.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Water transports heat from one site to another, so models
of temperature change in a watershed must address transport
and routing of water. Theurer and others (1985) used a
stream temperature model to compare the effects of potential
rehabilitation plans on stream temperatures in the Tucannon
River, Washington. The Timber/Fish/Wildlife Temperature
Work Group compared the sensitivity and accuracy of seven
predictive models for channel reaches and networks in areas
of timber management in Washington (Sullivan and others
1990), and recommended use of stream-reach models
described by Adams and Sullivan (1990) and Beschta and
Weatherred (1984; model up-dated 1986).
Research Needed to Understand
Watershed Processes
Land use alters environmental parameters, which then influ-
ence watershed processes. A watershed process is consid-
ered adequately understood for management purposes if
successful predictive models have been developed for it.
The development of such models requires an understanding
of the interactions between environmental parameters and
watershed processes.
Because water is the transport vehicle for watershed
products, an understanding of water transport through
watersheds is essential for understanding and predicting
transport of sediment, organic matter, chemicals, and heat.
Runoff generation processes are relatively well understood,
but models for predicting rain-on-snow runoff are needed to
evaluate the influence of logging pattern on flow generation.
Generation of surface flow on shallow saturated soils during
snowmelt is more common than previously thought, and the
runoff mode must therefore be reevaluated in some areas.
The effects of vegetation management on water yield are
widely measured, and these studies provide a data set for
developing and testing process-based predictive models.
Surface water transport on hillslopes can be approximately
modeled using existing methods. Subsurface flow, however,
requires further work to determine the significance and
influence of inhomogeneities in soil and bedrock. Flow
through subsurface soil pipes is common but rarely recognized,
and the influence of land-use activities on this flow path is
virtually unknown. Flow in alluvial channels is well
understood, and flow-routing models are capable of predicting
runoff timing and volume at points along a channel. Analogous
transport equations and models for high-gradient, low-order
streams are inadequate, yet these are often the channels of
most concern to foresters.
Most erosion processes have been widely studied, but
erosion in soil pipes and stream-bank erosion remain difficult
to evaluate. Landsliding has often been measured as a function
of land-use activity, and slide mechanics are fairly well
understood. However, predictions of long-term landsliding
79
rates are tenuous, and work is needed to evaluate how
occurrence of a landslide affects future erosion processes in
its vicinity.
Transport rates due to soil creep are inadequately measured,
and the effects of land-use activities on creep rates are
unknown. A change in creep rate can affect future sediment
input and process distribution, so definition of land-use-
related changes is important. The influence of altered creep
rate on landsliding and stream-bank erosion requires
evaluation.
The Universal Soil Loss Equation is effective for evaluating
sheetwash erosion on uncompacted lands. It is also useful
for identifying factors influencing erosion rates on compacted
surfaces and for estimating relative rates between sites, but
its numerical predictions are often not accurate for compacted
surfaces, and predictive equations for this application need
to be developed. Work is now being done to incorporate
road-surface erosion into the WEPP model, so appropriate
tools might be available in the near future.
Sediment transport in stream channels has been widely
studied and is relatively well-understood for low-gradient
alluvial channels. However, transport through high-gradient
channels with riffle-pool sequences and large organic debris
is not well described, and no predictive models are yet
adequate for this application. More detailed measurements
are needed to develop transport relations at such sites.
New types of sediment budgets are being explored that
allow prediction of landform evolution and spatially explicit
responses to environmental change (e.g., Reid 1989). Methods
80
for constructing these spatially distributed sediment budgets
need to be refined.
Input rates of organic debris into streams are easily
measured, and data exist for many areas. However, rates of
debris decay and breakdown are more poorly known, and
residence times must be determined as a function of species,
stream order, microhabitat, and hydrologic regime.
Measurement of organic material budgets for a variety of
site types would facilitate recognition of patterns of input
and transport and promote understanding of the factors
controlling them.
A variety of models now exist to predict transport and
accumulation of introduced chemicals, but these generally
apply to low-gradient, lowland settings. More work is needed
on movement of forest herbicides and pesticides so that
models may be adjusted for these species and environments.
Transport through inhomogeneous soils and steep, rough
waterways needs further research, and basic descriptive work
must still be done to determine distribution and residence
times of applied chemicals in various microenvironments.
Methods of evaluating nutrient cycles also need to be improved
so that sites at risk of nutrient depletion can be identified.
Models for predicting temperature change due to land use
and vegetation character in discrete stream reaches now
exist, but those that attempt to evaluate entire watersheds are
not as well developed. Most research on heat transport in
streams has considered summertime warming, and relatively
little is known of cold-season effects.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Chapter 7
Impacts on Beneficial Uses and Values
A change in environmental parameters induces
compensating changes in watershed processes, and a
combination of changes may produce responses that interact.
A CWE occurs when interacting responses are strong enough
to disturb something of philosophical or economic value,
and the significance of the impact is determined by the
magnitude of disturbance to the value. The potential effects
of watershed alterations on specific resources and values
must be understood if CWEs are to be evaluated, and this
chapter examines those effects. Many land-use activities
alter watershed function in the same way, so resource
responses -the impacts -are here related to the watershed
processes generating them rather than to the underlying
land-use activity. Almost all beneficial uses and values can
be affected by changes occurring in watersheds, and those of
particular concern in California are considered here.
Impacts on Fisheries Resources
and Aquatic Communities
Impacts on anadromous fish are often the focus of CWE
analyses in the western United States, and concerns over
impacts on resident fish are growing. Resident fish are af-
fected by changes in flow, physical habitat, temperature,
accessibility of food, predation, and pathogens within a wa-
tershed. Anadromous fish are influenced, in addition, by
changes occurring in habitats they use beyond their natal
watershed.
An impact on fisheries resources is usually recognized by
a change either in community structure or in abundance of a
target species. However, natural abundance is often difficult
to establish. Estimates of undisturbed populations have been
made on the basis of carrying capacities in streams and from
historical evidence (e.g., Chapman 1986), but large natural
population fluctuations can make evaluation of change difficult
(Platts and Nelson 1988).
Several publications reviewed potential impacts on
fisheries of particular land-use activities, such as hydropower
development (Russell and St. Pierre 1987), recreation (Clark
and others 1985), range use (Platts 1981), and timber
management (Chamberlin 1982, Chapman 1962, Meehan
and others 1969). Gibbons and Salo (1973) and MacDonald
and others (1988) provided comprehensive bibliographies
of the effects of timber management on fish, and Cordone
and Kelley (1961) reviewed impacts caused by increased
sediment load.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Flow Characteristics
Land use can change peak discharge, discharge variance,
low-flow discharge, and seasonal runoff distribution. Each
of these changes affects fish and the organisms they rely on
for food. Every species of fish has particular environmental
requirements for spawning, rearing, and adult stages. Salmo-
nids spawn in gravels by excavating a hole, depositing eggs,
and replacing gravel over the redd, so high flows during
incubation and intergravel life stages can scour redds and
wash away eggs or young fish. Natural floods periodically
destroy redds, and Thorne and Ames (1987) demonstrated a
negative correlation between sockeye smolt production per
spawner in Washington’s Cedar River and peak daily dis-
charge during incubation. Salmon compensate for these di-
sasters by having multiple runs in a single system, by spreading
a single run over several months, and by excavating redds to
depths unlikely to scour. In addition, some anadromous fish
return during off-years, so if one year’s production is de-
stroyed by a storm, pioneers from other years will return to
reestablish that run. These coping mechanisms are no longer
adequate if peak flows are consistently raised due to land-
use activities. Many other types of fish spawn on particular
substrates or aquatic plants, and these, too, can be disturbed
by unseasonal or chronically high flows.
Erman and others (1988) found that rain-on-snow floods
cause abnormally severe scour because they are contained
between snowbanks; a particular discharge is thus deeper
than usual and exerts more force on the channel bed. Snow
accumulation depths increase where riparian vegetation has
been cleared, so preservation of buffer strips in snow country
may help to prevent excessive scour.
Decreased flow can also affect spawning. Anadromous
adults require sufficient discharge during runs to allow
upstream migration, and parts of a stream system may become
inaccessible if these flows are delayed or prevented by
diversion or impoundment. Channels may also dry out if
accumulations of coarse sediment increase channel-bed
permeability and allow more flow to percolate through the
substrate. Seasonal or periodic desiccation can reduce the
spawning habitat available for both anadromous and resident
fish populations. Discharge fluctuations below hydroelectric
dams can increase in-gravel mortality by luring some fish to
spawn in gravels submerged during power generation but
exposed during off-hours. Chapman and others (1986) found
that 18 percent of chinook redds on a bar downstream of a
dam on the Columbia River were located in areas subject to
subaerial exposure, although little damage appears to have
occurred because redds were deep.
High peak flows can displace fish and destroy their food
resources. Ottaway and Clarke (1981) measured susceptibility
81
to displacement by high flows as a function of the ages of
brown trout and Atlantic salmon in artificial channels, and
Elwood and Waters (1969) attributed decreased brook trout
growth after flooding in central Minnesota to decreased
invertebrate populations. Low or fluctuating flows restrict
living space, strand fish, and expose them to predation and
high temperatures, and complete desiccation either kills or
displaces fish. Michael (1989) correlated survival of juvenile
salmonids with summer low-flow discharges in a Washington
stream, and Gibson and Myers (1988) demonstrated the
same effect in a Newfoundland stream where low flows were
caused by freezing during winter. In contrast, increased low
flows can benefit species by increasing available habitat
(Johnson and Adams 1988).
Changes in flow regime affect some species more than
others and so are likely to alter the composition of in-stream
communities. Bain and others (1988) compared fish
communities in natural flow regimes with those subjected to
artificially fluctuating flows and found that community
complexity was reduced by fluctuating flow.
Channel Morphology
Different species and life stages of fish have different habi-
tat requirements and react differently to changes. Gravel
spawners require a stable substrate while eggs and young
fish are resident in gravels, so morphological changes that
decrease substrate stability can reduce spawning success.
Land-use activities that rearrange stream gravels create a
bottom topography that is not adjusted to high flows. Rear-
ranged substrates are thus abnormally mobile during floods
as bed topography readjusts, and the increased scour can
destroy redds. Piles of gravel left by gold dredges provide
attractive sites for redds but are easily rearranged by high
flows. If few adult salmon escape to spawn and those that do
are lured to these unstable sites, the cumulative effect may
be a severe decrease in salmon populations.
Other activities affect substrate stability by altering the
size and distribution of stream gravels. Dietrich and others
(1989) found that the mean diameter of surface gravels
decreases as sediment supply increases, so increased sediment
input may make beds more mobile and increase the incidence
of redd scour. Debris flows destroy spawning and rearing
habitat in small tributaries by completely removing gravels,
so activities that destabilize slopes can strongly influence
spawning habitat.
Spawning gravel must be highly permeable so that
intergravel flow can deliver enough oxygen to support eggs
and young, and intergravel pore spaces must be large enough
that emerging fish can escape to the surface. Coble (1961)
and McNeil and Ahnell (1964) demonstrated decreased
survival to emergence for steelhead trout and pink salmon,
respectively, with increasing percent fines in gravels, and
Phillips (1970) reviewed evidence that suggests there is a
cumulative interaction between mortality due to decreased
intergravel flow and that due to blocking. Shaw and Maga
82
(1943) found that silver salmon emergence dates were delayed
when silt was added to gravels in a hatchery and suggested
that the delay might decrease fitness of fish during rearing
stages. Chapman (1988) reviewed research on the effects of
intergravel fines on salmonids.
The proportion of intergravel pore space occupied by fine
sediment increases with increased transport of fine sediment
(Lisle 1989). Meehan and Swanston (1977) found that angular
gravel accumulates more fines than rounded gravel, and
angular gravels contributed by landsliding may thus be more
susceptible to sedimentation than fluvially rounded gravels.
Increased sediment inputs from mining (Shaw and Maga
1943), roads (Sheridan and McNeil 1968), and logging
(Scrivener and Brownlee 1989) have been shown to increase
sedimentation in spawning gravels.
Rearing habitat must be stable during storms, give shelter
from predators, and provide appropriate flow velocities and
food resources. In particular, many species of fish require
pools for refuge, overhanging banks for cover and temperature
control, and organic debris for cover and food production
(Moore and Gregory 1988a). These qualities are largely
controlled by channel morphology, and fish move to more
suitable habitats if morphology changes. Salmonid species
often do not compete with each other because they prefer
different microhabitats. For example, steelhead fingerlings
are relatively common in riffles, while coho prefer pools
(Bisson and others 1988). Environmental changes that promote
one type of habitat at the expense of others are likely to alter
species composition in streams.
Many studies attempt to correlate fish populations to habitat
variables, and these were reviewed by Marcus and others
(1990). Kozel and others (1989) and Kozel and Hubert (1989),
for example, correlated the standing crop of brook trout and
brown trout to a variety of topographic and cover variables
for channels in meadows and forest lands in Wyoming. Moore
and Gregory (1988b) constructed complex banks in natural
reaches and found that cutthroat trout populations increased
with increasing area of channel margin habitat. Riparian
vegetation is a strong control on channel margin habitat and
food availability, and Dolloff (1987) demonstrated that
different densities of juvenile coho salmon were associated
with different riparian communities in southeast Alaska.
Other studies have isolated particular habitat elements
and have examined their influence on fish abundance. Morantz
and others (1987) showed that water velocity was a consistent
predictor of microhabitat selection for juvenile Atlantic salmon
and demonstrated that fish preferred higher water velocities
as they grew. Large gravel with open interstices provides
cover for juvenile trout (Heggenes 1988), but increased
sediment input destroys this habitat element by embedding
gravels in fine sediment. Some land-use activities reduce
cover by reducing inputs of organic debris to streams. Dolloff
(1986) and Elliott (1986) measured decreases in salmonid
size and population of juvenile Dolly Varden trout and coho
salmon when woody debris was removed from sites in
southeast Alaska, and Murphy and others (1986) found that
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
winter survival of juvenile salmonids in southeast Alaska
was correlated with the amount of woody debris left in
streams after logging. Bisson and others (1987) reviewed the
effects of woody debris on fish.
Habitat requirements vary with season. Site stability is
particularly important during winter, and some fish migrate
into floodplain channels, ponds, and backwater channels for
winter rearing. Young fish are protected from scouring flows
at these sites, and food is plentiful. Peterson (1982b) recorded
migration of coho into ponds from 33 km upstream, and
Peterson and Reid (1984) estimated that 20 to 25 percent of
smolt from the 375 km2 Clearwater drainage on the Olympic
Peninsula overwinters in ponds and tributaries that support
no spawning. Hartman and Brown (1987), at Carnation Creek
on Vancouver Island, and Swales and Levings (1989), in
interior British Columbia, examined differences in refuge
habitat use between species, and Brown and Hartman (1988)
found that 15 to 25 percent of the juvenile coho at Carnation
Creek migrated into off-channel refuges during the winter.
Peterson (1982a) measured survival and growth of juvenile
coho in two refuge ponds on the Olympic Peninsula and
attributed differences to contrasting pond morphology. Land-
use activities that degrade or make these habitats inaccessible
can disproportionately impact winter survival. Because these
sites are often on floodplains or low terraces, they are
commonly disturbed by agriculture, roads, urbanization, and
industrial development. Sedell and others (1990) discussed
the importance of refuge habitats of all scales for imparting
resistance and resilience to fish populations.
Fish also move to more protected sites within a reach
during winter, and their populations decrease if suitable
habitat is absent. Cunjak (1988) showed that Atlantic salmon
heavily used cobble substrates for winter refuge in Nova
Scotia, and Hillman and others (1987) increased overwinter
chinook population densities by introducing cobbles into
sedimented streams in Idaho. Tschaplinski and Hartman
(1983) examined winter population densities of coho in
Carnation Creek and found survival to be high in suitable
microhabitats both before and after logging. However, the
abundance of these microhabitats decreased after logging,
and more fish were displaced to off-channel refuges. Chapman
and Knudsen (1980) showed severely depressed winter
populations of coho in channels with morphology altered by
grazing or channelization in the Puget Sound area.
Some land-use activities alter morphology to improve
habitat for particular species, and others fortuitously improve
habitat. Cooper and Knight (1987) found that fish biomass
in plungepools created by erosion control structures was
comparable to that in natural pools in midwestern streams,
and that the growth rate, proportion of marketable fish, and
population stabilities were higher.
Channel obstructions, such as culverts, dams, and debris
jams, can eliminate fish migrations beyond those points, and
migrating fish can also be diverted by inlets to irrigation
channels, diversions, and turbines. In other cases, natural log
jams and waterfalls are blasted out to make new reaches
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
accessible, but such changes can impact recreational use and
esthetics and can increase sediment loads. Unique upstream
communities may be destroyed when removal of an
obstruction ends the isolation preserving them.
Sedimentation in streams is usually accompanied by
sedimentation and habitat loss in downstream estuaries.
Estuaries are extremely important as spawning grounds and
nurseries for coastal fish and invertebrates, and many
anadromous salmonids use estuaries during part of their life
cycle. Munro and others (1967, quoted in Gregory 1977)
documented destruction of an Australian oyster fishery by
sedimentation.
Water Temperature
Most fish thrive in a relatively narrow range of tempera-
tures, so temperature affects their choice of microhabitat.
Baltz and others (1987) found temperature, height above
bed, and flow depth to be the most important variables in
predicting the fish species present at sites along the Pit River
in California. Temperature affects fish behavior by altering
metabolic rates and demands, and juvenile chinook and coho
in a laboratory flume tended to stay in slower water at low
temperatures (Taylor 1988). Reeves and others (1987) com-
pared community composition in western Oregon streams of
different temperatures and found that different species domi-
nated at different temperatures, so temperature change is
likely to affect community composition. Beschta and others
(1987) reviewed the effects of temperature on stream biota.
Water temperature is moderated by shade, and prevention
of high temperatures is a primary reason for leaving buffer
strips in logged areas. Theurer and others (1985) used models
of fish population response to temperature change to predict
the effect of riparian revegetation on fish production in the
Tucannon River basin, Washington. Holtby (1988) related
changes in stream temperatures after logging in the Carnation
Creek watershed to fish production, and similar studies have
been carried out in other areas. Beschta and Taylor (1988)
demonstrated a cumulative increase in average maximum
temperature with increased logging disturbance in the Salmon
Creek watershed in the Oregon Cascades. Morphological
changes can also increase stream temperatures. Wide streams
are more quickly warmed than narrow streams with
comparable flow volumes because they have a large surface
area for temperature exchange and because marginal
vegetation shades less of the water volume.
Some management activities increase shading by providing
artificial structures or introducing debris. Meehan and others
(1987) found that biomass and abundance of juvenile chinook
salmon increased with the amount of shade added to sites
along a side channel of the South Fork Salmon River in
Idaho. Erman and others (1977) measured invertebrate
populations as a function of buffer strip width on 62 streams
in northwest California and found that 30-m buffers preserved
community structure.
Loss of canopy cover increases temperature fluctuations
83
by accelerating heat exchange, and fluctuations cause stress
in fish (Thomas and others 1986). Fluctuating discharges
from power generation also increase temperature variation,
and Hubbs (1972) reviewed the effects of impoundments on
water temperature and fish.
If streams are initially colder than optimal, artificial
warming can increase in-stream production, increase fish
populations, and accelerate growth. Bilby and Bisson (1987)
found greater coho production in clearcut channels than
under old-growth in the Deschutes River basin of western
Washington. Holtby (1988) showed a similar increase in
coho populations and attributed it primarily to earlier
emergence caused by higher early-season temperatures.
However, lowered ocean survival due to earlier out-migration
nearly compensated for the increased numbers. Power plants
add heat directly to streams when they return coolant water
to channels. Steen and Schubel (1986) warned of a tradeoff
between fish mortality due to warming and that due to
diversion: efforts to reduce thermal pollution require more
cooling water, so mortality from diversion increases.
Winter temperatures can also constrain survival. Fish
must have access to nonfreezing sites when ice forms.
Chisholm and others (1987) found that brook trout in high-
altitude streams in Wyoming migrated to low-gradient,
deep, low-velocity sites as streams started to freeze. Some
warming during cold months may increase growth rates,
but it can also increase food requirements during seasons of
low production.
Water Quality
Land-use activities affect water quality by altering sediment
and chemical loads, and both changes affect in-stream biota.
Norris and others (1983) reviewed the effects of forest pesti-
cides, fertilizers, and fire retardants on fish and suggested
that problems are most likely to arise when chemicals are
applied directly to surface waters. They stressed the impor-
tance of understanding the role and efficacy of buffer strips,
the dependence of toxicity on dose and duration, the instream
fate of chemicals, interactions between multiple chemicals,
and latent effects on biota.
High suspended sediment concentrations can harm animals
directly by damaging gills or influence them indirectly by
decreasing light penetration and visibility. Animals that hunt
by sight are less successful at gathering food in turbid water.
Bisson and Bilby (1982) found that juvenile coho attempted
to emigrate from water turbid enough to inhibit feeding, but
also showed that fish became somewhat habituated to elevated
turbidity. McLeay and others (1987) measured physiological
stress responses in grayling exposed to placer mining sediment
and observed direct mortality at high loadings. Chronic
exposure impaired feeding, decreased growth rates, and
reduced resistance to toxins. Reynolds and others (1989)
found that grayling that could not escape from sediment
plumes either died or sustained gill damage and arrested
growth. Redding and others (1987) observed physical stress
responses of coho salmon and steelhead trout to sublethal
84
sediment concentrations and documented a corresponding
reduction in tolerance to bacterial infection.
Fish may vary seasonally in their sensitivity to suspended
sediment as their metabolic demands and physiology change.
Noggle (1978, quoted in Cederholm and Reid 1987) found
that young Olympic Peninsula coho were most sensitive to
turbidity in summer and that tolerance increased during fall.
Servizi and Martens (1991) showed that temperature and
fish size are major controls on the sensitivity of coho salmon
to suspended sediments. Timing of sediment input may
therefore influence impacts, and high turbidities produced
by winter storms may be less damaging than equivalent
turbidities caused by dredging and road-building during the
summer. Fish from river systems with high natural suspended
sediment loads, such as the Eel River, may have evolved
physiological adaptations or coping behaviors that allow
them to tolerate higher sediment loads than fish inhabiting
less turbid streams.
Land-use activities can introduce many chemicals to
streams. Toxic spills often kill fish (e.g., Saunders 1969),
and coastal oil spills are relatively frequent. Siewert and
others (1989) measured changes in aquatic communities
after a toxic waste cleanup and found that species diversity
increased significantly when water quality improved. Toxic
chemicals introduced by mine drainage and leaching of spoils
can make channels uninhabitable (Platts and others 1979),
and Temmink and others (1989) described physiological
damage to fish from leaching of bark tannins around paper
mills. Hughs and Gammon (1987) circumstantially related
changes in fish assemblages along the Willamette River to
variations in water quality by demonstrating that an index of
biotic integrity decreased with decreasing water quality.
Toxic residues from marine anti-fouling paints have
accumulated around marinas, and Bushong and others (1988)
measured their effects on biota. Other chemicals are introduced
to streams to control vegetation or pests.
Chemical toxicities are usually determined in laboratories
rather than in the field, and results are often related only to
the lethal dose. Woodward (1978), however, found that
cutthroat trout exhibited a range of effects from reduced
growth to mortality as exposure to the herbicide picloram
increased, and Lorz and others (1979) measured some
physiological effects of atrazine on coho salmon even at
very low doses. Davis (1976) reviewed sublethal effects of
pollutants on salmonids and cited evidence for altered growth,
behavior, metabolism, respiration, and circulation, and also
noted that pollutants reduced tolerance of fish to temperature
extremes and made some game fish inedible. Kleerekoper
(1976) examined behavioral responses of fish to pollutants.
Heavy metal contamination and pollution in urban and
industrial areas have eliminated some local fisheries where
fish accumulate pollutants by eating contaminated food.
Ocean fisheries can also be affected by pollutants, as was the
case with mercury-tainted tuna.
Changes that reduce dissolved oxygen levels decrease
habitability for gilled creatures. Increased nutrient loads
deplete dissolved oxygen by stimulating algal blooms, and
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
detrital organic materials consume oxygen as they decay.
Dominy (1973) attributed fish kills below a Canadian dam
to supersaturation of nitrogen, which induced gas-bubble
disease in fish.
Some chemical changes interact to modify their effects
on fish. Aluminum toxicity decreases with silica content in
acidified streams (Birchall and others 1989), while the
combined effects of acidity and aluminum are worse than
either alone (Cleveland and others 1986). Increased silt load
decreases toxicity of some chemicals by adsorbing compounds
or altering their chemical environment (Hall and others 1986).
Woodward (1982) examined interactions between range
herbicides to assess combined toxicity to fish.
Food Resources
Most fish rely on plants, invertebrates, or other fish for food.
Foods can be produced within a channel reach, fall in from
banks, or drift in from upstream. Salmonids depend on both
instream and riparian food sources, so land use affecting
either infall or instream production alters food resources and
can affect salmonid populations. Infall depends primarily on
the nature of riparian vegetation, while instream production
is influenced by temperature, sedimentation, turbidity, nutrient
load, and channel morphology. Huryn and Wallace (1987)
found that invertebrate community composition could be
predicted by channel morphology in the southern Appalachians.
Wilzbach and others (1986) documented an increase in
insect drift in logged areas in the Oregon Cascades, and also
showed that decreased riparian shading and blocking of
interclast crevices aided cutthroat trout in catching their
prey. Jones and Clark (1987) showed that insect communities
decreased in complexity and species richness as intensity of
urbanization increased at sites in Virginia. Radford and
Hartland-Rowe (1972) found that in-stream invertebrate
biomass decreased with flow fluctuations below a reservoir
in Alberta, and Scrimgeour and Winterbourn (1989)
documented decreased algae and invertebrate populations
after flooding and gravel transport in a New Zealand river.
Increased turbidity and deposition of fine sediment on
streambeds also affect food resources. Sedimentation on
channel substrates reduces primary production by providing
unstable substrates and by decreasing light penetration. Barko
and Smart (1986) found that macrophyte growth is limited
on sandy or organic-rich sediments. Van Nieuwenhuyse and
LaPerriere (1986) demonstrated 50 percent decreases in
primary production downstream of moderate in-stream placer
mining in Alaska, and absence of primary production
downstream of heavily mined sites. Other studies have
measured the effects of siltation on aquatic invertebrate
communities (e.g., Luedtke and Brusven 1976).
Predation and Fishing Pressure
Fish are subject to natural predation, predation by intro-
duced species, and capture by sport, subsistence, and com-
mercial fisheries, but little information on predation rates
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
exists. Wood (1987b) measured predation rates of 24 to 65
percent on juvenile salmonids by mergansers in Vancouver
Island streams, but juvenile losses to other fish, aquatic
garter snakes, otters, kingfishers, and other birds were not
known. Wood (1987a) also showed that only 10 percent of
out-migrating smolt were consumed by mergansers in the
same system. Peterson (1982a) suggested that most over-
winter mortality in two off-channel ponds of the Olympic
Peninsula is by avian predation.
Natural predation is affected by altered vegetation and
channel morphology. Clearing of riparian vegetation can
both destroy predator habitat and remove cover used by fish
to escape predation. Shallowed channels, in-filled pools,
sedimented gravels and decreased flows can also reduce
cover and increase susceptibility to predation, and the effects
of these influences are often combined. Murphy and Hall
(1981) examined the effects of timber management on
predatory insects and vertebrates in streams of various sizes
in the Oregon Cascades and found that abundance increased
soon after logging but decreased when second-growth shades
channels. Russell and St. Pierre (1987) noted that
impoundments can increase predation on juvenile fish by
creating habitat for piscivorous fish.
Introduced predators can change the entire structure of a
community. Introduction of lamprey to the Great Lakes
destroyed productive fisheries, for example, and there is a
concerted effort to prevent introduction of piranhas to
California streams for fear of their effect on native fish.
Sport fisheries are closely regulated, and creel surveys
provide estimates of catch rates. Subsistence fisheries rarely
have catch records, however, and commercial ocean fisheries
involve so many nations that record-keeping is difficult.
Most information on ocean survival of salmonids is based on
the proportion of marked smolt recaptured as returning adults,
and these data combine the effects of natural predation and
fishing. Starr and Hilborn (1988) used information on migration
path, travel speed, and catch and escapement to estimate
catches by various fisheries for several stocks in the Pacific
Northwest. Fishing pressure is increasing as ocean fishing
technology becomes more sophisticated and recreational
fisheries grow. Current efforts to maintain sufficient stocking
levels depend primarily on regulation of coastal and sport
fisheries by establishing season dates and closed areas, while
the larger international fishery is largely unregulated.
River systems originally contained genetically distinct
stocks of fish adapted to local stream conditions (e.g., Bartley
and Gall 1990), and different strains could often be recognized
by size, color, timing of runs, and behavioral differences
(e.g., Rosenau and McPhail 1987). As native stocks declined,
they were often replaced by hatchery-raised fish derived
from only a few strains, and mixing of hatchery stock with
native stock further decreases regional genetic diversities
(Gausen and Moen 1991, Wehrhahn and Powell 1987).
With large hatchery releases, fishing pressure grows. If
hatchery and native fish are caught at the same rate and
hatchery stock outnumber native fish, lower absolute numbers
of native fish escape. Even though the total population may
85
survive a high catch-to-escapement ratio (C-to-E), this may
leave too few fish of a particular genetic stock to be viable.
In other words, a native stock may be able to support a 4:1
C-to-E. With a 3-fold population increase by addition of
hatchery stock, fishing pressure may rise to a 12:1 C-to-E
without changing the number of escaping fish. However, the
native stock also sustains the 12:1 C-to-E, and this represents
a 3-fold decrease in the number of native fish that escape.
Such pressure may extinguish genetically distinct strains
that evolved to suit particular conditions.
Michael (1989) found decreasing populations of sea-run
cutthroat in Olympic Peninsula creeks and suggested that
fishing pressure is contributing to their replacement by
nonmigratory cutthroat. Continuation of this trend will decrease
genetic diversity within the channel systems and is likely to
decrease the species’ ability to cope with changing
environmental conditions.
Other Effects
Nonnative game fish have been introduced in many areas,
and the effects of their competition with native species are not
well understood (Fausch 1988). Planting of hatchery stock
introduces pathogens and parasites contracted under crowded
hatchery conditions. Hatchery fish have introduced Bacterial
Kidney Disease (BKD) to some anadromous fish populations
in California, and the disease is now endemic in those
populations because it is transferable through eggs. Although
BKD does not kill fish directly, it severely reduces their
physiological resistance to stress. Infected fish thus become
more susceptible to environmental impacts, and resulting
mortality represents a cumulative effect of hatchery policy
and environmental change (J. Nielsen, USDA Forest Service
Pacific Southwest Research Station, personal communication).
Similarly, Servizi and Martens (1991) showed that a viral
kidney infection reduces tolerance of coho salmon to suspended
sediment in southern British Columbia.
Many anadromous fish depend on estuaries during part of
their life cycle, and estuaries are being severely altered by
land use. Reclamation projects fill in wetlands for building
sites, and accelerated upland erosion enhances sedimentation.
Reid and Trexler (1991) cited the cumulative effects of
dams, dredging, oil extraction, soil conservation, and
channelization on loss of wetland habitat and potential damage
to fisheries in Louisiana, and noted that a sea-level change
due to global warming would add an additional stress to this
environment. Wissmar and Simenstad (1988) evaluated
metabolic energy costs and intake rates to develop a model
of salmonid growth rates in estuaries and suggested that this
type of energy budget may be useful for assessing carrying
capacities for estuaries.
Barton and others (1986) demonstrated that the stress response
of salmonids to multiple handling disturbances is cumulative. If
this is a general response to environmental stress, then multiple
disturbances of any kind might produce cumulative physiological
effects and decrease fitness and survival.
86
Impacts on Timber Resources
Timber resources are adversely affected by activities that
restrict the area suitable for growing trees, decrease produc-
tivity of timberlands, or increase mortality of desired trees.
Because trees grow slowly, even very gradual changes can
be quite important.
Loss of Land Base
Competing land uses adversely affect timber resources by
decreasing the area available for logging and by restricting
the types and intensity of timber management activities.
Expansion of agriculture and urbanization reduce forested
area on private lands, while set-asides for reservoirs, wilder-
ness areas, and other uses incompatible with timber produc-
tion decrease the timber base on public land. Past timber
management has reduced forestland in some areas by inef-
fectual reforestation; some once-productive land now sup-
ports only successional hardwoods and chaparral.
Productive land base is also decreased by particular
activities. Logging roads and landings no longer produce
trees, and in some areas these account for 10 percent of the
surface (e.g., Miller and Sirois 1986, Ruth 1967). Elsewhere,
soil is completely removed by landslides and gullies and will
require decades to recover. Trustrum and DeRose (1988)
measured soil formation rates averaging 2.5 mm/yr on landslide
scars on deforested New Zealand slopes. Cumulative loss of
land base by combinations of these effects can be large.
Forest Productivity and Mortality
“Productivity” reflects the growth rates of desired species.
Compaction reduces growth rates, and as much as 27 percent
of a tractor-yarded site may be compacted (Dyrness 1965).
Duffy and McClurkin (1974) found that regeneration failure
on logged sites in northern Mississippi is best predicted by
soil bulk density, and Wert and Thomas (1981) showed that
compacted sites produced only 26 percent of the wood vol-
ume of noncompacted sites 32 years after logging in the
Oregon Coast Range.
Growth may also be slowed by loss of organic horizons,
erosion of mineral soils, and cumulative loss of nutrients.
Klock (1979) estimated changes in forest productivity from
logging-related soil erosion in the Washington Cascades.
Waide and Swank (1976) modeled nutrient balances over
three cutting cycles and predicted changes in forest productivity
for different cutting strategies in the southern Appalachians.
Routledge (1987) used economic models to demonstrate the
potential effects of such changes on timber values.
Factors that affect productivity may also influence tree
mortality. Thus, partial loss of a soil profile may increase
drought stress on seedlings and decrease survival rates. Misuse
of herbicides occasionally results in widespread seedling mortality
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
and foliar damage on treated slopes. Logging often provokes
increased populations of burrowing mammals that feed on
seedlings, and cattle may browse or trample young conifers if
grazing is improperly managed. Where channel sedimentation
rates are high, aggradation can raise water tables in riparian
zones and kill bottomland forests (Duda 1986).
Increased fire frequency and size can destroy wide areas
of forest, and such increases often accompany intensification
of land use. Some of the most destructive fires in California,
for example, have been caused by arson in urbanizing areas
and by recreational use. Disease and insects can also provoke
widespread mortality, and forest management practices may
increase susceptibility to these impacts. Logging equipment
can transfer pathogens to new sites (Zobel and others 1985),
and damage caused by thinning has been found to encourage
some insect pests (Mitchell and others 1983). Replacement
of mixed forests by monocultures may aid the spread of
some pests and diseases. Plantation forestry can reduce genetic
variation within a species in a region, and this may decrease
its resistance to some diseases, pests, and environmental
perturbations. Replacement of native strains by nursery-
bred stock may decrease fitness of a stock in particular
settings. Campbell (1979) found genetic differences between
native stands of trees growing in microhabitats a few tens of
meters from one another, and Fryer and Ledig (1972) showed
that genetically controlled photosynthetic temperature optima
for balsam fir closely followed the adiabatic lapse rate in the
White Mountains of New Hampshire. This type of genetic
specificity cannot be reproduced from nursery stock.
Air pollution has killed or damaged forests in many areas,
and acid rain is implicated in forest damage in central Europe
(e.g., Nihlgard 1985) and parts of North America. Pitelka
and Raynal (1989) reviewed the state of knowledge on acid
rain effects and discussed possible mechanisms for observed
forest changes. Toxic emissions have deforested wide areas
around Sudbury, Ontario (Amiro and Courtin 1981), and
have damaged forests around other smelter sites (e.g., Bruce
1989). Increased atmospheric ozone and other smog
constituents have damaged conifers in the Sierra Nevada
(Peterson and others 1987a) and decreased growth rates in
southern California (Omart and Williams 1979).
Large-scale climatic change from atmospheric pollution
has yet to be demonstrated, although many researchers expect
changes to become apparent within a few decades. If
atmospheric circulation patterns change, then temperature,
precipitation, and seasonality will be altered over broad areas.
Major vegetation changes are expected in areas like California,
where ecosystems intricately interfinger and are strongly
influenced by slight variations in climatic conditions. Ecotones
and forest types would be displaced along elevational gradients.
Predictions of climate change using global circulation models
have been used to estimate economic impacts of global warming
to timber income (e.g. Van Kooten and Arthur 1989).
Many forests in the eastern United States exhibit signs of
major decline, but reasons for the change have not been
determined. Hinrichsen (1986), Klein and Perkins (1987),
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
and Mazurski (1986) attributed forest decline in North America
and Europe to multiple and synergistic effects. They
hypothesized that combinations of air pollution and other
environmental impacts introduced multiple stresses into the
forest and made it more susceptible to mortality by drought,
insects, and disease. Hamburg and Cogbill (1988) suggested
that natural climatic warming is one cause of present forest
decline in the northeastern United States. Klein and Perkins
(1987) described a variety of effects that would be triggered
by forest decline and would further influence forest health.
Impacts on Recreation and
Esthetic Values
Recreational use and esthetic values strongly influence for-
est land-use planning because of the size of the recreational
constituency and because of its political importance. Tim-
ber-producing lands are often sites of heavy recreational
use, and the two interests are often in conflict; recreationists
are thus an important source of complaints about on-site
timberland conditions and forest practices. Sierra Club, Friends
of the River, Ducks Unlimited, Trout Unlimited, the Wilder-
ness Society, and many other environmental and conserva-
tion organizations are founded on both recreational and philo-
sophical values.
Recreational Needs
Different recreational pursuits require sites of different physi-
cal character, and recreationists also have less-well-defined
expectations that make some sites particularly attractive.
Satisfaction with a recreational experience depends strongly
on how well those expectations are met. Backpackers, for
example, are dissatisfied if areas are not pristine or if camp-
sites are overcrowded. If expectations in an area are consis-
tently unmet, then recreationists will shift their activities to
more desirable sites, their expectations will change, or they
will discontinue their activity. With the exception of watch-
ing television, participation levels for activities are directly
correlated to the level of satisfaction received (Ragheb 1980).
The attributes valued by particular types of recreationists
must be identified if the potential for impacts is to be
understood. Several studies explore recreationists’ motives
for participation and identify site characteristics that support
those motivations. Crandall (1980) identified general
motivations for use of leisure time, while Knopp (1972)
contrasted the needs of rural and urban groups and emphasized
that different populations have different recreational needs.
McDonald and Hammitt (1983) surveyed river rafters and
tubers and found that affiliation, experiencing nature, and
excitement were primary motives for their participation. On
this basis, the authors suggested that group-use sites be
developed, and that an element of risk be allowed. Searle
87
and Jackson (1985) identified barriers to recreational
participation and discussed their implications for recreation
management. Pfister (1977) evaluated reasons for selection
of particular campsites.
Within each recreational activity there are usually
subgroups with distinct motivations and values. Heywood
(1987) defined expectations of different types of river runners
and noted that “pick-up” groups are active primarily for
adventure and social contact, while parties of acquaintances
tend to participate for quiet escape. In addition, expectations
and standards change as a participant’s experience level
grows. Schreyer and others (1984) found that experienced
river runners are more critical of environmental degradation
and user conflicts than novice paddlers are.
Cumulative impacts on recreational use generally take
the form of decreased accessibility to appropriate sites, and
may involve changes in physical site characteristics, crowding,
and ambiance.
Physical Site Characteristics
Every recreational wildland pursuit requires particular site
conditions, and sites appropriate for some uses are rare. Few
sites are suitable for activities like whitewater rafting or
drift-boat fishing, and land use can further limit their
availability. Access to river recreation, for example, is
restricted by dam construction, hydropower diversions,
irrigation projects, and stream-side developments. A land-
use activity can affect a single recreational use in different
ways. Although dam construction destroys river recreation
sites, flow regulation may lengthen the recreation season
downstream: dam releases make some whitewater runs
boatable well into the dry season. Meanwhile, lake fishing
opportunities are created, while fluctuating flows may harm
downstream river fisheries. Recreational activities also have
basic environmental requirements for health and safety, and
physical changes brought about by land-use activities can
affect recreational risk.
Land use that alters flow affects recreational use of rivers.
Activities that decrease dry-season flows usually reduce
recreational opportunities, while those that enhance low flows
tend to promote recreation (Daubert and Young 1981). In
contrast, increases in higher discharges may increase hazards
to swimming and stream-side uses, and increased peak flows
destabilize channels and decrease their esthetic attraction.
Fluctuating flows usually decrease the overall recreational
potential (Radford and Hartland-Rowe 1972).
Altered channel morphology also affects river recreation.
Landslides often modify adjacent rapids and usually make
them more challenging for whitewater enthusiasts. Other
increases in sediment input usually broaden and aggrade
rivers, and this shortens the season for river recreation,
degrades the quality of rapids, and fills in pools used by
game fish. Removal of channel obstructions to promote fish
migration may destroy attractive rapids, while introduction
of logs to enhance fish habitat can make whitewater runs too
88
dangerous for recreational use. Fluctuating flow levels and
decreased sediment transport due to closure of Glen Canyon
Dam have eroded beaches used as campsites by Grand Canyon
river trips, and flow fluctuations degrade the esthetic
experience of the trips. Decreased peak flows below Flaming
Gorge Reservoir have allowed accumulation of coarse
sediment from tributaries, and this has increased the severity
of rapids in Dinosaur National Monument (Graf 1980).
Woody debris presents a safety hazard for river recreation
because it can trap swimmers underwater, and obstacles such as
bridge footings and scrap metal also contribute to drownings
and injuries. In particular, weirs and low-head dams are notorious
for their record of drowning swimmers in inescapable “holes”
of recirculating water created by their overfalls. Recreational
risk is increased by land uses that contribute these features, and
by uses that are in direct conflict with recreation. During the
summer of 1991 the most dangerous hazard on the popular
Pigeon Point whitewater run of the Trinity River was a gold
dredge located at the base of a rapid, and fear of assault by illicit
agriculturalists has effectively closed parts of several National
Forest districts to recreational use.
Opportunities for lake-based recreation have generally
been enhanced by land-use changes: there are many more
lakes now than there were before water and power
development. But artificial control of water levels in natural
lakes introduces fluctuations that destroy lakeshore plant
communities, and these changes decrease their attraction to
campers and bird watchers and can also affect fish populations.
Many activities require site developments like trail building,
campsite construction, and clearing of ski runs. Increased
urbanization usually creates an increased demand for such
facilities and increases the number available to choose from,
but increased use often more than compensates for new
facilities. Many recreational activities benefit from increased
road access. Mountain bikers, ORV drivers, rock collectors,
and anglers are quick to use new logging roads, and land
managers often must use extraordinary measures to exclude
recreational use from areas with road access. Cross-country
skiers frequent snow-covered roads, and new roads open rivers
to whitewater sports.
Land use can affect the incidence of disease contracted
during recreational use. Recreational use apparently has
contributed to the spread of Giardia, which is now a ubiquitous
threat to users of natural water supplies. Fecal coliform
counts are high in areas of heavy recreational use, suburban
development, and grazing, and high counts imply the presence
of pathogens. Introduction of environmental pollutants has
also increased health risks for recreational users. Fish taken
from contaminated water can pass on toxicants to those
eating them, and swimmers contract skin ailments and
infections from polluted waters.
Crowding and Ambiance
Crowding reduces the recreational value of many activities,
and is a cumulative effect of the decreasing number of
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
appropriate wildland sites and the increasing number of
wildland recreationists. Recognition of this conflict has led
to regulation of recreational use at the most desirable sites.
Permits for private whitewater trips on the Grand Canyon
now require several years’ wait, and campsites in popular
national parks are booked well in advance. Many wildland
recreationists avoid physically appropriate sites simply be-
cause too many people use them (Anderson and Foster 1985).
Manning (1985) reviewed factors contributing to perceived
crowding in wildlands and noted that most are subjective;
perceptions of crowding seem to have less to do with use
intensity than with users’ motivations, preferences, and ex-
pectations. Backpackers’ desires for solitude increase with
experience level (Stewart and Carpenter 1989), but even tent
campers rate opportunity for solitude as a more important
factor in their satisfaction with a camping experience than
facility condition, social interactions, and opportunity for
challenge and achievement (Connelly 1987).
However, social affiliation is also an important dimension
of some recreational activities. Recreational vehicle (RV)
campers often are attracted to crowded sites because they
enjoy a sense of community with other RV campers. The
margin of Highway 101 south of Redwood National Park,
for example, collects several hundred summer RV residents.
In other cases, crowding is simply not a consideration. Anglers
regularly stand shoulder-to-shoulder during steelhead runs
on Pacific northwest streams. Each would prefer to be the
only one present, but company is tolerated because the fish
are the major attraction. Downhill skiers complain of crowding
on slopes, but the social milieu of a ski resort is part of its
advertised attraction.
Because recreationists who value solitude usually choose
sites that are relatively free of other land uses, most conflicts
over solitude involve competing recreationists. Although users
indulging in the same activity can contribute to crowding, the
perception of solitude often is based mostly on the absence of
conflicting activities. Competing recreationists are tolerant
of others insofar as they perceive them to be similar to
themselves: operators of powerboats tend to view canoeists
as fellow boaters, while canoeists see people in powerboats
as motorists (Adelman and others 1982). Recreational activities
also conflict if they increase the risk for other users, and
ORV use is rarely compatible with horseback riding. Some
permit programs are designed to avert such conflicts.
Powerboats are restricted to certain seasons on the Grand
Canyon, for example, and those for whom peace is an important
recreational goal schedule trips for other seasons. Jacob and
Schreyer (1980) reviewed the basis for recreational conflicts.
Many wildland recreationists value a pristine setting. They
desire a wilderness experience, and any intrusion of “civilization”
decreases their enjoyment. Introduction of roads or logging
usually excludes backpackers from an area even if its physical
attributes are appropriate and solitude is ensured, and popular
backpacking destinations may be abandoned by backpackers
and adopted by tent campers when roads reach them.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Each additional development in a watershed contributes
to the cumulative loss of land available for recreational
activities that require pristine settings. Even recreational
developments are considered intrusive by many wildland
recreationists (Benthem 1973), and recreational activities
can conflict if they affect the ambiance of a site. Backpackers
are even repelled by evidence of previous hikers, although
standards are modified according to how close an impacted
site is to a trailhead (Shelby and others 1988). Recreational
use is usually concentrated in a few areas even if the potential
recreational land base is large, and preferred sites are
disproportionately degraded (Frissel and Duncan 1965). This
concentration of activities heightens the perceived loss of
pristine ambiance.
The value of many recreational experiences is degraded
in proportion to the severity of visual impacts caused by
other land uses, so recreational enjoyment can often be
maintained if coexisting land-use activities are designed to
minimize visual impacts. A high proportion of hikers at a
site in Maryland did not realize that the area had been
clearcut in the past and reported enjoying their experience
there even though they disapproved of clearcutting (Becker
1983). Esthetic values are an important component of
recreational use, and several methods have been developed
for assessing the esthetic attraction of a landscape. Leopold
(1969) presented a method for evaluating rivers based on
their physical characteristics, and Hamill (1976, 1986)
critiqued evaluation methods and suggested other elements
to be included. Sanderson and others (1986) related the
esthetic attraction of rangelands to management practices
and showed that perceived attraction depends on the
backgrounds of the individuals viewing a landscape. Work
has also been done to assess the esthetic impact of different
forest management options (e.g., Benson 1981).
Economic modeling studies have been carried out to
attempt to reconcile timber and recreational interests by
designing logging rotations that maximize the sum of timber
and recreation values (e.g., Englin 1990). However, this
approach may undervalue recreational use by not accounting
for changing recreational demands during future generations,
when recreational opportunities will be constrained by the
cumulative effects of today’s management decisions.
Impacts on Water Supply and
Power Generation
How suitable a water source is for domestic, industrial, and
agricultural supply depends on its quality, its yield, its tim-
ing, and the cost of developing the resource and transporting
the water. Suitability for power generation is affected by the
same factors, and by the topographic position of the source.
89
Water Quality
Water quality is degraded by sediment, chemicals, heat, and
bacterial content. Duda (1986) cites sediment as the major
impact on water quality in the southeastern United States.
Turbidity increases treatment cost and wear on power-generating
turbines, and is increased by activities that introduce fine
sediment to streams. Most land uses increase sediment input,
so most combinations of activities contribute cumulatively to
decreased water quality. Forster and others (1987) calculated
that a 10 percent reduction in cropland soil erosion in Ohio
would reduce water treatment costs by 4 percent.
Water quality is also degraded by chemical inputs. Many
land uses introduce herbicides, pesticides, and oil products,
and concentrations of these pollutants are sometimes high
enough to prevent domestic use. Atmospheric pollutants can
degrade water quality when they are deposited by precipitation,
and toxic wastes have contaminated aquifers. Of 45,000
wells tested in California, 5500 showed elevated
concentrations of potentially harmful agricultural chemicals
(Bouwer 1990). Contamination of water supplies by herbicides
has long been an issue, but its significance has not yet been
resolved. Kimmins (1975) reviewed the effects of silvicultural
herbicides. Hoar and others (1986) presented evidence for
an association between agricultural pesticide use and non-
Hodgkin’s lymphoma, and Colton (1986) reviewed studies
on this association. In the cases reviewed, affected individuals
had physical contact with the chemicals, and water supplies
were not a primary source of contamination.
Water Quantity and Runoff Timing
Water supplies are strongly affected by changes in the amount
and timing of surface runoff. Land-use activities that alter
infiltration rates or vegetation cover can change the propor-
tion of precipitation that runs off, and most activities affect
both of these properties. Deforestation increases water yield
by decreasing evapotranspiration rates, and vegetation has
often been manipulated to augment water supply. Urbaniza-
tion also usually increases runoff because of decreased evapo-
transpiration, decreased permeability, and water imports.
Changes in water yield are usually proportional to the area
of watershed altered, but the magnitude of the change de-
pends on the time-distribution of precipitation, type of pre-
cipitation, soil type, vegetation, and channel characteristics.
Enough water yield experiments have been carried out (e.g.,
Bosch and Hewlett 1982, Ffolliott and others 1989) that
general patterns of increase and the factors controlling their
magnitudes should be discernible.
Changes in runoff timing can alter the suitability of a
water source for particular uses. Power generation and most
consumptive water uses require relatively even water input
through the year, though irrigation demands are highest
during the growing season. Increased seasonality of runoff
increases dependence on water storage to balance supply
and demand, and reservoirs receiving most their inflow over
a short period must be capable of storing larger volumes
than those for which inflow is distributed through the year.
90
A snowpack represents a large volume of stored water. If the
proportion of precipitation falling as snow changes, or if
snowmelt is accelerated, snowpack storage must be replaced
by reservoir storage to maintain storage capacity.
Altered runoff changes the requirements for water
development. Climatic change can either increase or decrease
the demand for irrigation water, for example, and a change
in runoff seasonality can prolong or shorten the period during
which irrigation is required. Seasonality may change because
of climate alterations or because altered vegetation affects
the proportion of runoff occurring as rapid snowmelt.
Water Storage
Increased sedimentation in reservoirs can severely abridge
their usable lifespan. For example, China’s Sanmexia
Reservoir had silted in to the point that it was unusable in 4
years; California’s Black Butte Reservoir is losing storage
volume to sedimentation at a rate of 8 percent a decade; and
capacity behind Dam Number 3 on the Ocoee River in North
Carolina has decreased 73 percent in 30 years (Reisner
1986). Storage in most of the Nation’s reservoirs will decrease
noticeably within a few centuries and many will become
unusable. New water sources will need to be developed, but
the most suitable dam locations have already been used.
Regional sedimentation rates can be estimated using
compendia of reservoir infill data (e.g., Dendy and Champion
1978). Brune (1953) outlined a method for calculating
reservoir trap efficiencies if storage volume, inflow rates,
and river sediment loads are known.
Impacts on Floodplain and
Channel Use
Fertile floodplain soils often support intensive agriculture
and livestock production, and the sites provide relatively
level settings for roads, railways, and towns. Floodplain use
is particularly intensive in lowland areas distant from log-
ging and other wildland activities, but even these lowland
sites can be affected by forest practices upstream.
A variety of in-stream structures and modifications are
associated with floodplain use. Roads and railways require
bridges, and agricultural developments commonly include
irrigation diversions, levees, and bank-protection works.
Riverside urban centers are often serviced by river traffic,
which requires construction of ports, docks, and locks. These
uses and structures can all be affected by changes in the
morphology and flow of the rivers they depend on, and
occasionally by changes in aquatic communities.
Stream Flow
Peak flows can be altered chronically, as by a change in
basin hydrology or vegetation, or catastrophically, as by a
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
dam failure or major landslide. Chronic changes can in-
crease flood frequencies and cause property damage, loss of
life, increased maintenance costs for transportation and com-
munication systems, and damage to levees and bank protec-
tion structures. Occasional anthropogenic catastrophic events,
such as the Teton and St. Francis dam failures, have resulted
in one-time floods of extremely large size. Baldewicz (1984)
reviewed the incidence of catastrophic dam failures. In-
creased debris loading can cause log jams that back up flow
and increase effective stages for a given discharge, and
channel aggradation can have the same effect.
Low-flow discharges can be decreased by consumptive
water use, impoundments, and diversions, and lowered flows
may impede river navigation and floodplain irrigation.
Decreased low flows usually cause draw-down of riparian
water tables and increase irrigation demands. Aggradation
of coarse sediment in channels can reduce surface flows as
water percolates through permeable gravels, and this also
restricts the use of surface water.
Channel Morphology
Bank erosion undermines and destroys channel-bank struc-
tures and roads and increases maintenance and construction
costs. Channels can either move incrementally by bank ero-
sion or shift catastrophically, as when temporary damming
by a log jam diverts flow into an erodible overflow channel.
Resulting channel shifts can remove valuable lands from
production and replace them with infertile gravel bars or
widened channels. Major channel shifts can destroy struc-
tures far from the original channel. Channel changes pro-
voked by gravel mining along Tujunga Wash in southern
California destroyed a length of freeway and several bridges
(Bull and Scott 1974).
Increased sediment loads can aggrade lowland channels,
increase the frequency of overbank flows, and force channels
to migrate more rapidly. As bed elevation increases, channel-
bed gravels may be deposited at sites previously reached
only by silts, and unworkable gravels may cover productive
floodplain soils. Aggradation can also raise local water tables
so that drains must be constructed to maintain agriculture.
Sedimentation at domestic and irrigation water intakes can
increase maintenance and water treatment costs, and
aggradation or incision around bridges can endanger bridge
footings and occasionally lead to collapse and fatalities (Duda
1986). Increased upland erosion also contributes to accelerated
sedimentation in estuaries and may force more frequent dredging
of navigation channels.
In contrast, channel incision lowers adjacent water tables
and can make gravity-feed irrigation impossible. Ancient
gully erosion episodes in the American Southwest may have
destroyed the economic base of Native American civilizations
dependent on floodplain irrigation. Reduction of sediment
load downstream of a dam caused the Carmel River to
incise, and the resulting lowering of streamside aquifers
killed riparian vegetation (Matthews and Kondolf 1987).
Increased debris loading and log jams are associated with
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
altered patterns of bank erosion, channel migration, and
flooding. Anecdotal reports of saw-log jams tearing out
bridges during floods are common in forested areas of the
western United States, and choking of culverts by logging
debris is a common mechanism for logging road failures.
Other Changes
Imported water weeds can restrict navigation. For example,
navigation in Lake Washington is seasonally hampered by
milfoil growth, and expensive clearing operations are neces-
sary. Dams obstruct river navigation, and riverside commu-
nities may lose a source of transportation if locks are not
constructed. Floodplain soils are easily contaminated by
pollutants, because many chemicals accumulate on silt par-
ticles that preferentially settle out there. Lewin and others
(1977) and Marron (1989) measured elevated concentra-
tions of toxic chemicals in floodplain deposits downstream
of mines.
Impacts on Other Resources and
Values
Most resources and values are influenced by basic water-
shed properties either as habitat elements or as physical
influences on processes. Human health is affected by changes
in water quality and distribution of toxic materials and patho-
gens. Changes in vegetation, erosion rates, and hydrology
affect range resources and wildlife, and cattle health has
been impaired by mining-related chemical changes (Erdman
and others 1978).
Progressive soil erosion alters soil structure and depletes
nutrients on croplands. Schertz and others (1989) measured
decreased soybean yields of 24 percent at sites with severe
cumulative soil erosion in Indiana. Water development in
combination with repeated use for irrigation increases salinity
in many rivers of the western United States, and may
eventually raise concentrations enough that further use is
impossible. Brown and others (1990) calculated the cost of
maintaining tolerable salt levels in the Colorado River.
Colacicco and others (1989) reviewed a variety of impacts
resulting from agricultural soil erosion and estimated costs
associated with them.
The western United States is rich with the cultural heritage
of Native Americans. Many natural areas and topographic
features hold spiritual significance, and the cumulative cultural
loss grows as incompatible uses encroach on an increasing
number of these sites. In addition, many sites of spiritual and
cultural importance are located along rivers and are susceptible
to destruction by altered flows and channel morphology.
Economists have defined several types of noneconomic
values that are affected by environmental changes. These
address the observation that many people value the existence
91
of healthy forests, wild rivers, undisturbed areas, and other
qualities of wilderness. In some cases, importance is attached
to preserving a quality for future generations, while in others
people want to know that the opportunity to experience a
setting is available for them if they should decide to visit it in
the future. People also recognize the importance of preserving
genetic resources and biodiversity in case future needs arise.
Methods of assessing these values are not yet satisfactory.
Pope and Jones (1990) attempted to quantify the value of
wilderness preservation in Utah by asking residents how
much they would pay to preserve wilderness. Walsh and
others (1990) and Sanders and others (1990) used similar
methods to measure the value of forest quality and river
protection, respectively. Results of such studies can be difficult
to interpret, however, because a significant proportion of the
respondents often refuse to consider the concept of paying
for something that they believe is a fundamental mission of
public land management. In the case of the study by Sanders
and others (1990), these respondents were simply excluded
from the analysis. Increasingly, environmental organizations
such as Nature Conservancy and Save-the-Redwoods are
demonstrating the economic value of land preservation by buying
land to exclude incompatible uses.
Research Needed to Understand
Impacts
Much basic research is needed to understand and predict
responses of aquatic communities to changing conditions.
Little is known of survival strategies during harsh winter
conditions or of mechanisms of recovery from disturbance.
92
Many impacts involve slight changes in timing of events,
and the effects of these changes on aquatic biota are only
now beginning to be studied. Interactions between compet-
ing organisms are poorly understood, and these have a strong
influence on the response of organisms and communities to
change. Habitat variables that control fish use need to be
better defined to improve the usefulness of habitat assess-
ments and to predict the effects of altered habitat. For ex-
ample, Baltz and others (1987) demonstrated that the three
habitat variables most often used for in-stream flow studies
were not optimal for predicting species use of microhabitats
in a northern California river.
Vegetation changes brought about by suspected systemic
changes in atmospheric composition and global climate may
be of increasing importance to timber resources but are very
poorly understood. All other land-use related changes to
forest productivity will be overlaid on this backdrop, and
effects will be difficult to isolate. Work is needed to quantify
the effects of ongoing, well-understood changes on second-
and third-cycle logging. In particular, impacts of altered
nutrient regimes are poorly understood, and the influence of
management strategies on the resilience and stability of
forest ecosystems must be determined.
Further work on recreational use patterns and needs is
necessary to determine the effects of altered access and
physical and biological setting on recreational values.
Motivations and needs must be determined for various activity
groups, and patterns of recreational response to changing
conditions need to be understood. Impacts of altered conditions
on recreational safety are not well known. Esthetic and other
noneconomic values must be far better understood if impacts
on these resources are to be evaluated adequately, and methods
to assess the magnitude of such impacts need to be developed.
Many of these problems require sociological and psychological
research.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Chapter 8
Summary of Research and Cumulative Watershed Effects

O ne researcher has described cumulative effects as “the

effects of everything on anything.” Although intended
to be flippant, this description underscores the inherent
complexity of the subject. Prediction of CWEs would be
essentially impossible without a framework for understanding
their nature and causes. A strictly empirical approach to
prediction cannot succeed because too many land-use activities
can combine in too many ways and affect too many potential
resources and values.
A Framework for Understanding
CWEs
The problem of understanding CWEs is simplified when
potential mechanisms for triggering impacts are considered.
Land-use activities can influence only a limited number of
environmental parameters (path A in Figure 5). Changes in
vegetation, soil characteristics, topography, chemicals, water,
pathogens, and fauna can induce compensatory changes in
one another (path B) and can influence watershed processes
(path C). Watershed processes arise from an area’s role as a
concentrator of runoff, and include production and transport
of runoff, sediment, chemicals, organic material, and heat.
These processes can influence environmental parameters (path
D), and they can also interact (path E). Changes in either
environmental parameters or watershed processes can generate
on-site CWEs (paths F and G), but only changes in watershed
processes can produce off-site CWEs (paths H and I).
A cumulative effect is a change influenced by multiple,
progressive, or repeated activities. Effects can accumulate
through time or grow by contributions from multiple sources.
Accumulations through time require that either the triggering
mechanism be persistent, or the recovery time be greater
than the period between disturbances. In contrast, a spatially
accumulating effect may occur through instantaneous changes.
Cumulative effects are complicated by properties of
watersheds and ecosystems that obscure the relation between
cause and effect. Effects may be delayed until long after the
triggering activity has occurred, and frequently the location
of impact is far removed from the original land-use disturbance.
Local conditions may modify the form of an impact, and a
single impact may have many contributing causes.
Some changes may accumulate until a threshold is reached
and then trigger a catastrophic change that is not evident until
it occurs. Thus, slow channel incision may progress until
banks are high enough to fail. Other changes may accumulate
benignly until the occurrence of an external triggering event
like a large storm. These effects also may not be evident until
after preventative measures are no longer possible. Threshold-
type changes are particularly insidious from a management
perspective, because precedents for their occurrence may not
exist. Such effects can be predicted only from a basic
understanding of the processes contributing to them.
State of Present Knowledge
The ability to predict CWEs depends on understanding how
and why the effects occur. Complete understanding would
require knowledge of all interactions represented in figure 5.
Many of these have been extensively studied, and some
interactions are now qualitatively or even quantitatively pre-
dictable. Most fields of natural science have contributed to
our present understanding of CWEs, and work in hydrology,
geomorphology, and ecology is particularly relevant.
Direct influences of specific land-use activities on
environmental parameters are easily observed and measured,
and abundant data exist. In particular, effects of most activities

F G D off-site H
on-site
environmental
impact
parameter

Land A on-site C I
watershed off-site
use environmental
process impact
activities parameter

B E

Figure 5—Influences that generate on-site and off-site CWEs. Interaction paths are described in the text.

USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993. 93

on vegetation, soil properties, topography, water import, and
faunal import are well-defined. Influences on chemical import
and pathogen distribution are more poorly understood, but
some measurements exist. The nature of interactions between
altered environmental parameters can often be predicted, but
quantitative information is rare. Land-use activities having
similar influences on an environmental parameter will have
similar influences on watershed processes, so once the nature
of the direct environmental change is defined, the type of
land-use generating it is no longer of importance to evaluation
of impacts.
Rates, distributions, and modes of watershed processes
are controlled by environmental parameters. The influence
of environmental parameters on production and transport of
water and sediment have been quantified in many areas, and
predictive equations exist for many processes and conditions.
Credible water and sediment budgets can now be developed
for most areas using available methods, published data, and
a few local measurements, and effects of land-use activities
on sediment production and runoff can thus be estimated.
Results are usually only approximate, but are adequate for
most analyses. Methods of flow-routing in channels are
well-established and allow estimation of changes in peak
flow due to changes in runoff distribution, but available
techniques are not well suited for rough, high-gradient
channels. Sediment routing is more poorly understood because
methods to evaluate sediment delivery are poorly developed.
Future input rates for organic debris often can be estimated
for planned land-use changes using local measurements of
mortality and edge effects. Decay and breakdown rates of
woody debris in channels are rarely known, however, and
too few measurements have been made to develop general
relationships. Qualitative patterns are evident and may be
predicted, but quantitative predictions of altered debris
transport and accumulation are not yet possible.
Components of nutrient cycles have been measured at
many sites, but nutrient production and transport processes
are not well understood. Both nutrient loss at a site and
nutrient accumulation downstream may cause impacts and
need to be predicted. Qualitative predictions of loss and
accumulation can often be made on the basis of local
experience and general patterns, but quantitative predictions
are not yet possible. Because of current concerns over toxic
waste disposal, methods are being developed for tracking
the movement of introduced chemicals through watersheds.
Results of these studies are applicable to the problem of
nutrient routing and downstream accumulation. Production
and transport of heat has been more widely modeled, and
several methods are now being used to predict the effects of
altered land use on stream temperatures.
Few predictive models for watershed processes can assess
the results of interactions between processes. For example, a
temperature model may accurately predict changes occurring
from increased flow, but the thermal impact of secondary
effects, such as that due to the loss of riparian vegetation
from destabilized streambanks, are not predicted by a flow-
94
based model. Similarly, changes in sediment input can alter
channel morphology, which may then modify rates of sediment
input from bank erosion. Predictive methods should not be
applied without evaluating the potential for secondary changes
that might affect results.
Our current inability to accurately calculate rates of
sediment accumulation and scour in channels as a function
of sediment input and flow is of particular concern. These
mechanisms are the driving force for altering channel
morphology, and morphological change affects other
watershed processes and frequently impacts resources. Patterns
and trends can often be predicted qualitatively, but quantitative
predictions are rarely possible.
The final steps in the framework illustrated by figure 5
(paths F, G, H, and I) are the links between altered parameters
or processes and resource impacts. In many cases, these
links are well known and predictable. For example, loss of
reservoir capacity is directly proportional to sedimentation,
and rates can be calculated from measured or predicted
sediment yields and trap-efficiency equations. In general,
impacts on physical resources can often be predicted if the
nature of changes in parameters or processes is known.
However, responses of biological systems are more difficult
to predict because of the complex interdependencies within
biological communities. Correlative studies have identified
some general patterns of response among communities, so a
population’s response to change can often be predicted
qualitatively. Quantitative predictions are possible only in
systems where the nature of biological interactions and
dependencies on physical habitat are understood.
Many studies have provided information on the knowledge
gaps identified above, and results of existing studies need to
be compiled and examined for patterns of response. This
approach is often useful for identifying variables that control
the expression of a response and for focusing future studies.
Methods for Evaluating CWEs
Land managers need usable and accurate procedures for
evaluating the combined effects of multiple land-use activi-
ties on other resources and values. Although several proce-
dures have been developed and are in use, none can be
applied generally. Current models were developed to apply
to particular issues in particular areas, and do not address the
variety of land-use activities, mechanisms of watershed re-
sponse, and types of impacts that occur at other sites.
Although a general predictive technology does not yet
exist, the ability to predict many types of effects clearly does,
and many studies have evaluated and predicted cumulative
effects of particular land-use activities on particular resources.
Successful studies usually evaluated impacts by examining
the processes that are altered by land use, or by comparing
multiple sites to infer patterns and magnitudes of occurrence.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
 Three general approaches to predictive CWE assessment
have been developed:
1. Models that produce a predicted value for a physical or
biological parameter or an index value that indicates the
potential for triggering a particular type of impact.
2. A collection of procedures for use in evaluating a
variety of types of impacts, where a relevant subset is selected
for a particular application.
3. A checklist of items to consider during assessment, but
with specific approaches left to be selected by the operator.
The first approach may work well in the area for which a
model was developed, but the model cannot be applied to
sites with different characteristics, processes, or types of
impact. Most of these models assess the potential for an
impact at a specific time and do not take into account the
extent to which the impact has occurred in the past. Because
of this, potential impacts that accumulate because of their
slow recovery rate cannot be assessed. The second approach
allows the flexibility of determining the impacts and
mechanisms relevant to a particular setting, but provides no
guidance for deciding which aspects must be evaluated. The
final approach provides the guidance but not the methods.
A Framework for CWE Evaluation
A workable general procedure for CWE evaluation might
contain elements of all three approaches. A checklist, key, or
expert system could guide users through a decision tree to
identify the types of impacts likely to be of concern in a
management region. Appropriate procedures could then be
selected from a compilation of state-of-the-art methods to
evaluate identified concerns. These procedures might in-
clude existing methods from categories 1 and 2.
This combined approach has the advantage of flexibility in
addressing unique concerns and characteristics of particular
sites while still providing standardized evaluation procedures.
In addition, it allows improved predictive methods to be
incorporated as they become available, and thus supports use of
the best available technology for each aspect evaluated. This is
a critical requirement both for responsible land management
and for accountability in the face of litigation, and it recognizes
a fundamental characteristic of technology: no product is
definitive for long. Approaches that incorporate a capacity to
evolve will be useful far longer than those that do not.
In practice, a particular management region would be
characterized by a relatively small subset of likely impacts
and corresponding evaluations. This subset could be
incorporated into a streamlined “cookbook” for use in the
region. Implementation would be relatively straightforward
after measurement of the coefficients and relationships
required by the selected procedures. Any implementation of
a CWE assessment method must include a mechanism for
evaluating its validity.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
The Role of Research in
Assessing CWEs
Research is essential in all phases of development and
implementation of a CWE evaluation procedure. Model
development depends on research that defines relationships
between change and response or explains the mechanisms by
which response occurs. This requires continued basic research
to increase understanding of how watersheds and biological
communities function. It also requires increased emphasis on
the knowledge gaps discussed above, and interdisciplinary
work is necessary if interactions between physical and
biological systems are to be clearly understood and predicted.
Once an impact mechanism is identified and understood,
that knowledge must be transformed into an ability to predict
response. At this point, interaction between researchers,
resource specialists, and land managers is essential to ensure
that the procedures developed are useful and practical. Models
that are useful to managers provide results at scales relevant
to management problems, are not overly vague or overly
detailed in their predictions, represent the range of conditions
likely to occur in the management setting, and do not require
excessive data for calibration and use.
Each impact mechanism requires a model or evaluation
procedure, and each model or procedure will be outdated as
further research reveals additional information. Research to
understand and predict CWEs will thus continue as long as
CWEs remain an issue.
Research may be necessary to provide data for model
calibration. Most models require an initial calibration for
application to particular settings. This may simply involve
measurement of topographic attributes of a region, or it may
require construction of a sediment budget to assess sediment
inputs for various types of land use in the region.
Verification is essential for the confident application of
any model or procedure. In the case of CWE assessments,
verification is likely to include two independent aspects.
First, the validity of individual modules must be tested by
comparing predicted and measured values. This must be
done both during model development and under field
conditions. Verification usually requires monitoring of
selected sites.
The second type of verification considers the success of
the CWE assessment as a whole, rather than that of its
components. To accomplish this, all impacts generated at a
selection of evaluated sites must be surveyed and compared
to the risks predicted by the original CWE assessments.
This, too, will require input from researchers on experimental
design and monitoring techniques.
Methods are currently available for qualitatively or
quantitatively predicting many types of CWEs from many
types of land use, so initial development of a protocol for
CWE assessment need not wait for new research results.
95
Appendix—People Interviewed
During Report Preparation

Name Affiliation Location

Neil Berg USDA Forest Service, Pacific Southwest Research Station Albany, Calif.
C. Jeffery Cederholm Washington State Department of Natural Resources Olympia, Wash.
John Chatoian USDA Forest Service, Region 5 San Francisco, Calif.
Glenn Chen USDA Forest Service, Siskiyou National Forest Powers, Oregon
Lynn Decker USDA Forest Service, Region 5 San Francisco, Calif.
William Dietrich University of California,Department of Geology and Geophysics Berkeley, Calif.
Thomas Dunne University of Washington, Department of Geological Sciences Seattle, Wash.
Frederick Euphrat University of California, Department of Forestry Berkeley, Calif.
Fred Everest USDA Forest Service, Pacific Northwest Research Station Juneau, Alaska
Michael Furniss USDA Forest Service, Six Rivers National Forest Eureka, Calif.
Gordon Grant USDA Forest Service, Pacific Northwest Research Station Corvallis, Oregon
Russell Henly University of California, Department of Forestry Berkeley, Calif.
Dennis Harr USDA Forest Service, Pacific Northwest Research Station Seattle, Wash.
Eugene Hetherington Forestry Canada Victoria, British
James Hornbeck USDA Forest Service - Northeastern Research Station Durham, N.H.
George Ice National Council of the Paper Industry for Air and Stream Improvement Corvallis, Oreg.
Terry Kaplan-Henry USDA Forest Service, Sequoia National Forest Porterville, Calif.
Harvey Kelsey Western Washington University, Department of Geology Bellingham, Wash.
G. Mathias Kondolf University of California, Department of Landscape Architecture Berkeley, Calif.
Antonius Laenen US Geological Survey Portland, Oregon
George Leavesly US Geological Survey Denver, Colo.
Donna Lindquist Pacific Gas and Electric Company San Ramon, Calif.
Thomas Lisle USDA Forest Service, Pacific Southwest Research Station Arcata, Calif.
Keith Loague University of California, Soil Science Department Berkeley, Calif.
Lee MacDonald Colorado State University, Department of Earth Resources Fort Collins, Colo.
Mary Ann Madej Redwood National Park Arcata, Calif.
Bruce McGurk Southwest Research Station Albany, Calif.
Walter Megahan National Council of the Paper Industry for Air and Stream Improvement Port Townsend, Wash.
David Montgomery University of Washington, Department of Geological Sciences Seattle, Wash.
Thomas Nickelson Oregon Department of Fish and Wildlife Corvallis, Oreg.
Jennifer Nielsen USDA Forest Service, Pacific Southwest Research Station Arcata, Calif.
K. Michael Nolan US Geological Survey Menlo Park, Calif.
N. Phil Peterson University of Washington Center for Streamside Studies Seattle, Wash.
Gordon Reeves USDA Forest Service, Pacific Northwest Research Station Corvallis, Oreg.
Ray Rice USDA Forest Service, Pacific Southwest Research Station Arcata, Calif.
Roy Sidle USDA Forest Service, Intermountain Research Station Logan, Utah
John Stednik Colorado State University Department of Earth Resources Fort Collins, Colo.
Kathleen Sullivan Weyerhaeuser Company Tacoma, Wash.
Fred Swanson USDA Forest Service, Pacific Northwest Research Station Corvallis, Oreg.
William Trush Humboldt State University Fisheries Department Arcata, Calif.
Clyde Wahrhaftig US Geological Survey Menlo Park, Calif.
Robert Wissmar University of Washington, Center for Streamside Studies Seattle, Wash.
Peter Wohlgemuth USDA Forest Service, Pacific Southwest Research Station Riverside, Calif.
Robert Ziemer USDA Forest Service, Pacific Southwest Research Station Arcata, Calif.

96 USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.


References
Abbott, P.O. 1976. Observed channel changes in a mountain stream
due to increased flow from transbasin imports. In: Proceedings of
the third federal inter-agency sedimentation conference; 1976 March
22-25; Denver, CO. Washington, DC: Water Resources Council,
Sedimentation Committee; 5-25 to 5-36.
Abdel-Magid, A.H.; Schuman, G.E.; Hart, R.H. 1987a. Soil bulk density
and water infiltration as affected by grazing systems. Journal of
Range Management 40(4): 307-309.
Abdel-Magid, A.H.; Trlica, M.J.; Hart, R.H. 1987b. Soil and vegetation
responses to simulated trampling. Journal of Range Management
40(4): 303-306.
Abrahams, A.D.; Luk, S.-H.; Parsons, A.J. 1988. Threshold relations for
the transport of sediment by overland flow on desert hillslopes.
Earth Surface Processes and Landforms 13(5): 407-420.
Abrams, M.D.; Scott, M.L. 1989. Disturbance-mediated accelerated
succession in two Michigan forest types. Forest Science 35(1): 42-49.
Adams, J.N.; Beschta, R.L. 1980. Gravel bed composition in Oregon
coastal streams. Canadian Journal of Fisheries and Aquatic Sciences
37: 1514-1521.
Adams, R.M.; Rowe, R.D. 1990. The economic effects of stratospheric
ozone depletion on U.S. agriculture: a preliminary assessment.
Journal of Environmental Management 30(4): 321-335.
Adams, T.A.; Sullivan, K. 1990. The physics of forest stream heating: a
simple model. Timber/Fish/Wildlife Report TFW-WQ3-90-007.
Olympia, WA: Washington Department of Natural Resources.
Adelman, B.J.E.; Heberlein, T.A.; Bonnickson, T.M. 1982. Social
psychological explanations for the persistence of conflict between
paddling canoeists and motorcraft users in the Boundary Waters
Canoe Area. Leisure Sciences 5(1): 45-61.
Ahnert, F. 1976. Brief description of a comprehensive three-dimensional
process-response model of landform development. Zeitschrift für
Geomorphologie Neue Folge Supplementband 25: 29-49.
Alberts, E.E.; Moldenhauer, W.C. 1981. Nitrogen and phosphorus
transported by eroded soil aggregates. Soil Science Society of America
Journal 45(2): 391-395.
Alexander, R.B.; Smith, R.A. 1988. Trends in lead concentrations in
major U.S. rivers and their relation to historical changes in gasoline-
lead consumption. Water Resources Bulletin 24(3): 557-569.
Amaranthus, M.P.; Rice, R.M; Barr, N.R; Ziemer, R.R. 1985. Logging
and forest roads related to increased debris slides in southwestern
Oregon. Journal of Forestry 83: 229-233.
Amiro, B.D.; Courtin, G.M. 1981. Patterns of vegetation in the vicinity
of an industrially disturbed ecosystem, Sudbury, Ontario. Canadian
Journal of Botany 59: 1623-1639.
Anderson, B.; Potts, D.F. 1987. Suspended sediment and turbidity
following road construction and logging in western Montana. Water
Resources Bulletin 23(4): 681-690.
Anderson, D.C.; Harper, K.T.; Rushforth, S.R. 1982. Recovery of
cryptogamic crusts from grazing on Utah winter ranges. Journal of
Range Management 5(3): 355-359.
Anderson, D.H.; Foster, D.I. 1985. Perceived change in a river
environment and its effect on visitor use: a case study of Ozark
National Scenic Riverways. Western Wildlands 11(2): 21-24.
Anderson, H.W. 1954. Suspended sediment discharge as related to
stream flow, topography, soil and land use. Transactions, American
Geophysical Union 35: 268-281.
Anderson, H.W.; Hobba, R.L. 1959. Forests and floods in the northwestern
United States. Publ. no. 49. International Association of Scientific
Hydrology; 30-39.
Anderson, S.H.; Gantzer, C.J.; Brown, J.R. 1990. Soil physical properties
after 100 years of continuous cultivation. Journal of Soil and Water
Conservation 45(1): 117-120.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Andrus, C.W.; Long, B.A.; Froehlich, H.A. 1988. Woody debris and its
contribution to pool formation in a coastal stream 50 years after
logging. Canadian Journal of Fisheries and Aquatic Sciences 45(12):
2080-2086.
Anthony, D.J.; Harvey, M.D. 1987. Response of bed topography to
increased bed load, Fall River, Colorado. In: Erosion and sedimentation
in the Pacific Rim. Publ. no. 165. International Association of
Hydrological Sciences; 387-388.
Armour, C.D.; Bunting, S.C.; Neuenschwander, L.F. 1984. Fire intensity
effects on the understory in ponderosa pine forests. Journal of
Range Management 37(1): 44-49.
Armson, K.A.; Fessenden, R.J. 1973. Forest windthrow and their influence
on soil morphology. Soil Science Society of America Proceedings 37:
781-783.
Arno, S.F.; Gruell, G.E. 1983. Fire history at the forest grassland
ecotone in southwestern Montana. Journal of Range Management
36(3): 332-336.
Arno, S.F.; Sneck, K.M. 1977. A method for determining fire history in
coniferous forests of the mountain west. Gen. Tech. Rep. INT-42.
Ogden, UT: Intermountain Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 28 p.
Arnold, J.G.; Bircket, M.D.; Williams, J.R.; Smith, W.F.; McGill, H.N.
1987. Modeling the effects of urbanization on basin water yield and
reservoir sedimentation. Water Resources Bulletin 23(6): 1101-1107.
Aubertin, G.M.; Patric, J.H. 1974. Water quality after clearcutting a
small watershed in West Virginia. Journal of Environmental Quality
3(3): 243-249.
Austin, D.D.; Urness, P.J.; Riggs, R.A. 1986. Vegetal change in the
absence of livestock grazing, mountain brush zone, Utah. Journal of
Range Management 39(6): 514-517.
Austin, R.D.; Baisinger, D.H. 1955. Some effects of burning on forest
soils of western Oregon and Washington. Journal of Forestry 53:
275-280.
Bailey, J.; Hobbs, V. 1990. A proposed framework and database for
EIA auditing. Journal of Environmental Management 31(2): 163-172.
Bain, M.B.; Finn, J.T.; Booke, H.E. 1988. Streamflow regulation and
fish community structure. Ecology 69(2): 382-392.
Baker, T.G; Will, G.M.; Oliver, G.R. 1989. Nutrient release from
silvicultural slash: leaching and decomposition of Pinus radiata
needles. Forest Ecology and Management 27(1): 53-60.
Baldewicz, W.L. 1984. Dam failures: insights to nuclear power risks.
In: Waller, R.A.; Covello, V.T., eds. Low-probability/high-consequence
risk analysis. New York: Plenum Press; 81-92.
Balph, D.F.; Malecheck, J.C. 1985. Cattle trampling of crested wheatgrass
under short-duration grazing. Journal of Range Management 38(3):
226-227.
Baltz, D.M.; Vondracek, B.; Brown, L.R.; Moyle, P.B. 1987. Influence of
temperature on microhabitat choice by fishes in a California stream.
Transactions of the American Fisheries Society 116: 12-20.
Barko, J.W.; Smart, R.M. 1986. Sediment-related mechanisms of growth
limitation in submersed macrophytes. Ecology 67(5): 1328-1340.
Barr, D.J.; Swanston, D.N. 1970. Measurement of creep in shallow,
slide-prone till soil. American Journal of Science 269: 467-480.
Bartley, D.M.; Gall, G.A.E. 1990. Genetic structure and gene flow in
chinook salmon populations of California. Transactions of the
American Fisheries Society 119(1): 55-71.
Barton, B.A.; Schreck, C.B.; Sigismondi, L.A. 1986. Multiple acute
disturbances evoke cumulative physiological stress responses in
juvenile chinook salmon. Transactions of the American Fisheries
Society 115(2): 245-251.
Barton, M.A. 1969. Water pollution in remote recreational areas. Journal
of Soil and Water Conservation 24: 132-134.
Beanlands, G.E.; Erckmann, W.J.; Orians, G.H.; O’Riordan, J.; Policansky,
D.; Sadar, M.H.; Sadler, B., eds. 1986. Cumulative environmental
effects: a binational perspective. Canadian Environmental Assessment
Research Council; 175 p.
Beasley, R.S. 1979. Intensive site preparation and sediment losses on
steep watersheds in the Gulf Coastal Plain. Soil Science Society of
America Journal 43: 412-417.
97
Beasley, R.S.; Granillo. A.B. 1988. Sediment and water yields from
managed forests on flat coastal plain sites. Water Resources Bulletin
24(2): 361-366.
Beaton, J.D. 1959. The influence of burning on the soil in the timber
range area of Lac Le Jeune, BC: I. Physical properties. Canadian
Journal of Soil Science 39: 1-5.
Becker, R.H. 1983. Opinions about clear-cutting and recognition of
clearcuts by forest recreation visitors. Journal of Environmental
Management 17(2): 171-177.
Bell, K.L.; Bliss, L.C. 1973. Alpine disturbance studies: Olympic National
Park. Biological Conservation 5(1): 25-32.
Bell, R.W.; Schofield, N.J.; Loh, L.C.; Bari, M.A. 1990. Groundwater
response to reforestation in the Darling Range of western Australia.
Journal of Hydrology 115: 297-317.
Belt, G.H. 1980. Predicting streamflow changes caused by forest practices
using the equivalent clearcut area model. Publ. no. 32. Moscow:
University of Idaho Forest, Wildlife and Range Experiment Station; 11 p.
Bencala, K.E.; Walters, R.A. 1983. Simulation of solute transport in a
mountain pool-and-riffle stream: a transient storage model. Water
Resources Research 19(3): 718-724.
Benda, L.E. 1985. Delineation of channels susceptible to debris flows
and debris floods. In: Proceedings of the international symposium on
erosion, debris flow and disaster prevention; 1985 Sept 3-5; Tsukuba,
Japan; 195-201.
Benda, L.; Dunne, T. 1987. Sediment routing by debris flows. In: Erosion
and sedimentation in the Pacific Rim. Publ. no. 165. International
Association of Hydrological Sciences; 213-224.
Bengston, G.W. 1981. Nutrient conservation in forestry: a perspective.
Southern Journal of Applied Forestry 5: 50-58.
Benson, M.A. 1965. Spurious correlation in hydraulics and hydrology.
Journal of the American Society of Civil Engineers, Hydraulics Division
91(HY4): 35-42.
Benson, R.E. 1981. Visual impacts of forest management activities:
findings on public preferences. Res. Paper INT-263. Ogden, UT:
Intermountain Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 14 p.
Benson, V.W.; Rice, O.W.; Dyke, P.T.; Williams, J.R.; Jones, C.A. 1989.
Conservation impacts on crop productivity for the life of a soil.
Journal of Soil and Water Conservation 44(6): 600-604.
Benthem, R.J. 1973. Recreation and environmental planning. Biological
Conservation 5(1): 1-5.
Bergman, D.L.; Sullivan, C.W. 1963. Channel changes on Sandstone
Creek near Cheyenne, Oklahoma. 475C. Washington, DC: Geological
Survey, U.S. Department of Interior; 145-148.
Berndt, H.W.; Swank, G.W. 1970. Forest land use and streamflow in
central Oregon. Res. Paper PNW-93. Portland, OR: Pacific Northwest
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture; 15 p.
Berndtsson, R. 1990. Transport and sedimentation of pollutants in a
river reach: a chemical mass balance approach. Water Resources
Research 26(7): 1549-1558.
Berris, S.N.; Harr, R.D. 1987. Comparative snow accumulation and
melt during rainfall in forested and clear-cut plots in the western
Cascades of Oregon. Water Resources Research 23(1): 135-142.
Berry, L.; Townshend, J. 1972. Soil conservation policies in the semi-
arid regions of Tanzania, a historical perspective. Geografiska Annaler
54A(3-4): 241-253.
Beschta, R.L. 1978. Long term patterns of sediment production following
road construction and logging in the Oregon Coast Range. Water
Resources Research 14(6): 1011-1016.
Beschta, R.L. 1990. Effects of fire on water quantity and quality. In:
Natural and prescribed fire in Pacific Northwest forests. Corvallis, OR:
Oregon State University Press; 219-232.
Beschta, R.L.; Bilby, R.E.; Brown, G.W.; Holtby, L.B.; Hofstra, T.D.
1987. Stream temperature and aquatic habitat: fisheries and forestry
interactions. In: Salo, E.O.; Cundy, T.W., eds. Streamside management:
forestry and fishery interactions. 57. Seattle, WA: University of
Washington Institute of Forest Resources; 191-232.
98
Beschta, R.L.; Jackson, W.L. 1979. The intrusion of fine sediments into
a stable gravel bed. Journal of the Fisheries Research Board of Canada
36(2): 204-210.
Beschta, R.L.; Taylor, R.L. 1988. Stream temperature increases and
land use in a forested Oregon watershed. Water Resources Bulletin
24(1): 19-26.
Beschta, R.L.; Weatherred, J. 1984. TEMP-84. A computer model for
predicting stream temperatures resulting from the management of
streamside vegetation. WSDG-AD-00009. Fort Collins, CO: Watershed
Systems Development Group. Available from U.S. Department of
Agriculture, Washington, DC.
Bethlahmy, N. 1962. First year effects of timber removal on soil moisture.
Bulletin of the International Association of Scientific Hydrology 7: 34-38.
Bethlahmy, N. 1967. Effect of exposure and logging on runoff and
erosion. Res. Paper INT-61. Ogden, UT: Intermountain Forest and
Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 7 p.
Bethlahmy, N.; Kidd, W.J. 1966. Controlling soil movement from steep
road fills. Res. Note INT-45. Ogden, UT: Intermountain Forest and
Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 4 p.
Bethlenfalvay, G.J.; Dakessian, S. 1984. Grazing effects on mycorrhizal
colonization and floristic composition of the vegetation on a semi-
arid range in northern Nevada. Journal of Range Management 37(4):
312-316.
Bilby, R.E. 1981. Role of organic debris dams in regulating the export
of dissolved and particulate matter from a forested watershed.
Ecology 62: 1234-1243.
Bilby, R.E.; Bisson, P.A. 1987. Emigration and production of hatchery
coho salmon (Oncorhynchus kisutch) stocked in streams draining
an old-growth and a clear-cut watershed. Canadian Journal of Fisheries
and Aquatic Sciences 44(8): 1397-1407.
Bilby, R.E.; Sullivan, K.; Duncan, S.H. 1989. The generation and fate of
road-surface sediment in forested watersheds in southwestern
Washington. Forest Science 35(2): 453-468.
Bilby, R.E.; Ward, J.W. 1989. Changes in characteristics and function
of woody debris with increasing size of streams in western
Washington. Transactions of the American Fisheries Society 118(4):
368-379.
Birchall, J.D.; Exley, E.; Chappell, J.S.; Phillips, M.J. 1989. Acute toxicity
of aluminum to fish eliminated in silicon-rich acid waters. Nature
338(6211): 146-147.
Bishop, D.M.; Stevens, M.E. 1964. Landslides on logged areas in southeast
Alaska. Res. Paper NOR-1. St. Paul, MN: Northern Forest Experiment
Station, Forest Service, U.S. Department of Agriculture; 18 p.
Bisson, P.A.; Bilby, R.E. 1982. Avoidance of suspended sediment by
juvenile coho salmon. North American Journal of Fisheries Management
2(4): 371-374.
Bisson, P.A.; Bilby, R.E.; Bryant, M.D.; Dolloff, C.A.; Grette, G.B.;
House, R.A.; Murphy, M.L.; Koski, K.V.; Sedell, J.R. 1987. Large
woody debris in forested streams in the Pacific Northwest: past,
present, and future. In: Salo, E.O.; Cundy, T.W., eds. Streamside
management: forestry and fishery interactions. 57. Seattle, WA:
University of Washington Institute of Forest Resources; 143-191.
Bisson, P.A.; Sullivan, K.; Nielsen, J.L. 1988. Channel hydraulics, habitat
use, and body form of juvenile coho salmon, steelhead, and cutthroat
trout in streams. Transactions of the American Fisheries Society
117(3): 262-273.
Biswell, H.H.; Schultz, A.M. 1957. Surface run-off and erosion as related
to prescribed burning. Journal of Forestry 55: 372-374.
Bjerklie, D.M.; LaPerriere, J.D. 1985. Gold-mining effects on stream
hydrology and water quality, Circle quadrangle, Alaska. Water
Resources Bulletin 21(2): 235-243.
Blackburn, W.H.; Knight, R.W.; Wood, M.K.; Merrill, L.B. 1980.
Watershed parameters as influenced by grazing. In: Symposium on
watershed management 1980. New York: American Society of Civil
Engineers; 552-569.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Blackburn, W.H.; Wood, J.C.; DeHaven, M.G. 1986. Storm flow and
sediment losses from site-prepared forestland in East Texas. Water
Resources Research 22(5): 776-784.
Blake, G.R.; Nelson, W.W.; Allmaras, R.R. 1976. Persistence of subsoil
compaction in a mollisol. Soil Science Society of America Journal 40:
943-948.
Boden, R.W.; Ovington, J.D. 1973. Recreation use-patterns and their
implications for management of conservation areas. Biological
Conservation 5(4): 265-270.
Boe, K.N. 1965. Windfall after experimental cuttings in old-growth
redwood. In: Proceedings of the Society of American Foresters, Detroit,
MI. Society of American Foresters; 59-63.
Bohn, C.C.; Buckhouse, J. 1985. Some responses of riparian soils to
grazing management in northeastern Oregon. Journal of Range
Management 38(4): 378-381.
Bolton, H., Jr.; Smith, J.L.; Wildung, R.E. 1990. Nitrogen mineralization
potentials of shrub-steppe soils with different disturbance histories.
Soil Science Society of America Journal 54(3): 887-891.
Bosch, J.M.; Hewlett, J.D. 1982. A review of catchment experiments to
determine the effect of vegetation changes on water yield and
evapotranspiration. Journal of Hydrology 55: 3-23.
Bouwer, H. 1990. Agricultural chemicals and groundwater quality.
Journal of Soil and Water Conservation 45(2): 184-189.
Braden, J.B.; Herrick, E.E.; Larson, R.S. 1989. Economic targeting of
nonpoint pollution abatement for fish habitat protection. Water
Resources Research 25(12): 2399-2407.
Bradford, J.M.; Piest, R.F. 1977. Gully wall stability in loess-derived
alluvium. Soil Science Society of America Journal 41(1): 115-122.
Branson, F.A. 1985. Vegetation changes on western rangelands. Range
Monograph no. 2. Denver, CO: Society for Range Management; 76 p.
Bray, D.I. 1987. A study of channel changes in a reach of the North
Nashwaaksis Stream, New Brunswick, Canada. Earth Surface
Processes and Landforms 12(2): 151-165.
Bren, L.J.; Leitch, C.J. 1985. Hydrologic effects of a stretch of forest
road. Australian Forest Research 15(2): 183-194.
Brockway, D.G. 1983. Forest floor, soil, and vegetation responses to
sludge fertilization in red and white pine plantations. Soil Science
Society of America Journal 47(4): 776-784.
Brookes, A. 1987. River channel adjustments downstream from
channelization works in England and Wales. Earth Surface Processes
and Landforms 12(4): 337-352.
Brown, G.W. 1970. Predicting the effect of clearcutting on stream
temperature. Journal of Soil and Water Conservation 25: 11-13.
Brown, G.W.; Krygier, J.T. 1971. Clear-cut logging and sediment
production in the Oregon Coast Range. Water Resources Research
7(5): 1189-1198.
Brown, J.C.; Evans, R.A.; Young, J.A. 1985. Effects of sagebrush control
methods and seeding on runoff and erosion. Journal of Range
Management 38(3): 195-199.
Brown, T.C.; Harding, B.L.; Payton, E.A. 1990. Marginal economic
value of streamflow: a case study for the Colorado River basin.
Water Resources Research 26(12): 2845-2860.
Brown, T.G.; Hartman, G.F. 1988. Contribution of seasonally flooded
lands and minor tributaries to the production of coho salmon in
Carnation Creek, British Columbia. Transactions of the American
Fisheries Society 117(6): 546-551.
Bruce, H.W.F. 1989. The continuing effects of aluminium smelter
emissions on coniferous forest growth. Forestry 62(3): 223-231.
Brune, G.M. 1953. Trap efficiency of reservoirs. Transactions, American
Geophysical Union 34(3): 407-418.
Bryan, R.B. 1979. The influence of slope angle on soil entrainment by
sheetwash and rainsplash. Earth Surface Processes 4(1): 43-58.
Bryant, M.D. 1980. Evolution of large organic debris after timber
harvest: Maybeso Creek, 1949 to 1978. Gen. Tech. Rep. PNW-101.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 30 p.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Buckhouse, J.C.; Gifford, G.F. 1976. Water quality implications of cattle
grazing on a semiarid watershed in southeastern Utah. Journal of
Range Management 29: 109-113.
Bull, W.B.; Scott. K.M. 1974. Impact of mining gravel from urban
stream beds in the southwestern United States. Geology 2: 171-174.
Burcham, L.T. 1982. California range land. Publ. no. 7. Davis, CA:
Center for Archaeological Research at Davis; 256 p.
Burger, J.A.; Pritchett, W.L. 1988. Site preparation effects on soil moisture
and available nutrients in a pine plantation in the Florida flatwoods.
Forest Science 34(1): 77-87.
Burkham, D.E. 1976. Hydraulic effects of changes in bottom-land
vegetation on three major floods, Gila River in southeastern Arizona.
Professional Paper 655-J. Washington, DC: Geological Survey, U.S.
Department of Interior; 13 p.
Burns, J.W. 1970. Spawning bed sedimentation studies in northern
California streams. California Fish and Game 56(4): 253-270.
Burroughs, E.R.; King, J.G. 1989. Reduction of soil erosion on forest
roads. Res. Paper INT-264. Ogden, UT: Intermountain Forest and
Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 21 p.
Burroughs, E.R., Jr.; Watts, F.J.; Haber, D.F. 1984. Surfacing to reduce
erosion of forest roads built in granitic soils. In: O’Loughlin, C.L.;
Pearce, A.J., eds. Proceedings of a symposium on effects of forest land
use on erosion and slope stability; 1984 May 7-11; Honolulu, HI.
Honolulu, HI: East-West Center, University of Hawaii; 255-264.
Bushong, S.J.; Hall, L.W., Jr.; Hall, W.S.; Johnson, W.E.; Herman, R.L.
1988. Acute toxicity of tributyltin to selected Chesapeake Bay fish
and invertebrates. Water Research 22(8): 1027-1032.
Buttle, J.M.; Xu, F. 1988. Snowmelt runoff in suburban environments.
Nordic Hydrology 19: 19-40.
Caine, N. 1980. The rainfall intensity-duration control of shallow
landslides and debris flows. Geografiska Annaler 62A(1-2): 23-27.
Caine, N.; Swanson, F.J. 1989. Geomorphic coupling of hillslope and
channel systems in two small mountain basins. Zeitschrift für
Geomorphologie N.F. 33(2): 189-203.
Callaham, R.Z. 1990. Guidelines for management of wildland watershed
projects. Publ. no. 23. Berkeley, CA: University of California, Wildland
Resources Center; 36 p.
Callaham, R.Z.; DeVries, J.J., eds. 1987. Proceedings of the California
watershed management conference; 1986 November 18-20; West
Sacramento. Publ. no. 11. Berkeley, CA: University of California,
Wildland Resources Center; 167 p.
Campbell, C.E. 1972. Some environmental effects of rural subdividing
in an arid area. Journal of Geography 71: 147-154.
Campbell, R.K. 1979. Genecology of Douglas-fir in a watershed in the
Oregon Cascades. Ecology 60: 1036-1050.
Carlson, J.Y.; Andrus, C.W.; Froehlich, H.A. 1990. Woody debris, channel
features, and macroinvertebrates of streams with logged and
undisturbed riparian timber in northeastern Oregon, USA. Canadian
Journal of Fisheries and Aquatic Sciences 47: 1103-1111.
Carr, W.W.; Ballard, T.M. 1980. Hydroseeding forest roadsides in British
Columbia for erosion control. Journal of Soil and Water Conservation
35(1): 33-35.
Carrighar, S. 1943. One day on beetle rock. Curtis Publication Co.
Carson, R.L. 1962. Silent spring. Boston, MA: Houghton Mifflin; 368 p.
Caspary, H.J. 1990. An ecohydrological framework for water yield
changes of forested catchments due to forest decline and soil
acidification. Water Resources Research 26(6): 1121-1133.
California Department of Forestry. 1991. Guidelines for assessment of
cumulative watershed effects. Sacramento, CA: California Department
of Forestry and Fire Protection.
Cederholm, C.J.; Peterson, N.P. 1985. The retention of coho salmon
(Oncorhynchus kisutch) carcasses by organic debris in small streams.
Canadian Journal of Fisheries and Aquatic Sciences 42: 1222-1225.
99
Cederholm, C.J.; Reid, L.M. 1987. Impact of forest management on
coho salmon (Oncorhynchus kisutch) populations of the Clearwater
River, Washington: a project summary. In: Salo, E.O.; Cundy, T.W.,
eds. Streamside management: forestry and fishery interactions. Publ.
no. 57. Seattle, WA: University of Washington Institute of Forest
Resources; 373-398.
Cerco, C.F. 1989. Measured and modelled effects of temperature,
dissolved oxygen, and nutrient concentration on sediment-water
nutrient exchage. Hydrobiologia 174(3): 185-194.
Chamberlin, T.W. 1982. Influence of forest and range land management
on anadromous fish habitat in western North America. 3. Timber
harvest. Gen. Tech. Rep. PNW-136. Ogden, UT: Intermountain Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture.
Chanasyk, D.S.; Verschuren, J.P. 1980. Hydrologic simulation of the
effects of forest removal. In: Symposium on watershed management
1980. New York: American Society of Civil Engineers; 452-464.
Chapman, D.W. 1962. Effects of logging upon fish resources of the
West Coast. Journal of Forestry 60: 533-537.
Chapman, D.W. 1986. Salmon and steelhead abundance in the Columbia
River in the nineteenth century. Transactions of the American Fisheries
Society 115(5): 662-670.
Chapman, D.W. 1988. Critical review of variables used to define effects
of fines in redds of large salmonids. Transactions of the American
Fisheries Society 117(1): 1-21.
Chapman, D.W.; Knudsen, E. 1980. Channelization and livestock impacts
on salmonid habitat and biomass in western Washington.
Transactions of the American Fisheries Society 109(4): 357-363.
Chapman, D.W.; Weitkamp, D.E.; Welsh, T.L.; Dell, M.B.; Schadt, T.H.
1986. Effects of river flow on the distribution of chinook salmon
redds. Transactions of the American Fisheries Society 115(4): 537-547.
Cheng, J.D. 1989. Streamflow changes after clear-cut logging of a pine
beetle-infested watershed in southern British Columbia, Canada.
Water Resources Research 25(3): 449-456.
Cheng, J.D.; Black, T.A.; de Vries, J.; Willington, R.P.; Goodell, B.C.
1975. The evaluation of initial changes in peak streamflow following
logging of a watershed on the West Coast of Canada. Publ. no. 117.
International Association of Scientific Hydrology; 475-486.
Chisholm, I.M.; Hubert, W.A.; Wesche, T.A. 1987. Winter stream
conditions and use of habitat by brook trout in high-elevation
Wyoming streams. Transactions of the American Fisheries Society
116(2): 176-184.
Chorley, R.J.; Kennedy, B.A. 1971. Physical geography - a systems
approach. London: Prentice-Hall, Inc.; 370 p.
Christensen, L.A.; McElyea, D.E. 1988. Toward a general method of
estimating productivity-soil depth response relationships. Journal
of Soil and Water Conservation 43(2): 199-203.
Christner, J.; Harr, R.D. 1982. Peak streamflows from the transient
snow zone, eastern Cascades, Oregon. In: Proceedings of the 50th
Annual Western Snow Conference; 27-38.
Church, M.; Slaymaker, O. 1989. Disequilibrium of Holocene sediment
yield in glaciated British Columbia. Nature 337(6206): 452-453.
Clark, R.N.; Gibbons, D.R.; Pauley, G.B. 1985. Influence of forest and
rangeland management on anadromous fish habitat in western
North America. 10. Influence of recreation. Gen. Tech. Report PNW-
178. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture.
Clark, T.; Euler, D. 1984. Vegetation disturbance caused by cottage
development in central Ontario. Journal of Environmental Management
18(3): 229-239.
Cleveland, L.; Little, E.E.; Hamilton, S.J.; Buckler, D.R.; Hunn, J.B. 1986.
Interactive toxicity of aluminum and acidity to early life stages of
brook trout. Transactions of the American Fisheries Society 115(4):
610-620.
Cline, R.; Cole, G.; Megahan, W.; Patten, R.; Potyondy, J. 1981. Guide for
predicting sediment yields from forested watersheds. Ogden, UT:
Intermountain Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 48 p.
100
Coats, R.N.; Miller, T.O.; Kallstrom, D.W. 1979. Assessing cumulative effects
of silvicultural activities. Napa, CA: John Muir Institute, Inc.; 160 p.
Coble, D.W. 1961. Influence of water exchange and dissolved oxygen in
redds on survival of steelhead trout embryos. Transactions of the
American Fisheries Society 90(4): 469-474.
Cobourn, J. 1989. An application of cumulative watershed effects (CWE)
analysis on the Eldorado National Forest in California. In:
Headwaters hydrology. Bethesda, MD: American Water Resources
Association; 449-468.
Cochran, P.H. 1968. Can thinning slash cause a nitrogen deficiency in
pumice soils of central Oregon? Res. Note PNW-82. Portland, OR:
Pacific Northwest Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 11 p.
Colacicco, D.; Osborn, T.; Alt, K. 1989. Economic damage from soil
erosion. Journal of Soil and Water Conservation 44(1): 35-39.
Cole, D.N. 1978. Estimating the susceptibility of wildland vegetation to
trailside alteration. Journal of Applied Ecology 15(1): 281-286.
Collier, C.R.; Whetstone, G.W.; Musser, J.J. 1964. Influences of strip
mining on the hydrologic environment of parts of Beaver Creek
Basin, Kentucky, 1955-1959. Professional Paper 427-C. Washington,
DC: U.S. Department of Interior, Geological Survey; 85 p.
Collins, B.D.; Dunne, T. 1989. Gravel transport, gravel harvesting, and
channel-bed degradation in rivers draining the southern Olympic
Mountains, Washington, U.S.A. Environmental Geology and Water
Sciences 13(3): 213-224.
Collins, S.L. 1987. Interaction of disturbances in a tallgrass prairie: a
field experiment. Ecology 68(5): 1243-1250.
Colton, T. 1986. Herbicide exposure and cancer. Journal of the American
Medical Association 256: 1176-1179.
Connelly, N.A. 1987. Critical factors and their threshold for camper
satisfaction at two campgrounds. Journal of Leisure Research 19(3):
159-173.
Cook, M.J.; King, J.G. 1983. Construction cost and erosion control
effectiveness of filter windrows on fill slopes. Res. Note INT-335.
Ogden, UT: Intermountain Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 5 p.
Cooper, A.C. 1965. The effects of transported stream sediments on the
survival of sockeye and pink salmon eggs and alevin. Publ. no. 18.
International Pacific Salmon Fisheries Commission; 71 p.
Cooper, C.M.; Knight, S.S. 1987. Fisheries in man-made pools below grade-
control structures and in naturally occurring scour holes of unstable
streams. Journal of Soil and Water Conservation 42(5): 370-373.
Cordone, A.J.; Kelley, D.W. 1961. The influence of inorganic sediment on
the aquatic life of streams. California Fish and Game 47(2): 189-228.
Costa, J.E. 1975. Effects of agriculture on erosion and sedimentation in
Piedmont province, Maryland. Geological Society of America Bulletin
86: 1281-1286.
Courtney, F.M. 1981. Developments in forest hydrology. Progress in
Physical Geography 5(2):217-241.
Cowan, G.I.M.; Baggs, E.M.; Hollohan, B.T. 1989. Bacteriological water
quality of a multi-use catchment basin on the Avalon Peninsula,
Newfoundland. Water Research 23(3): 261-266.
Crandall, R. 1980. Motivations for leisure. Journal of Leisure Research
12: 45-54.
Cryer, M.; Corbett, J.J.; Winterbotham, M.D. 1987. The deposition of
hazardous litter by anglers at coastal and inland fisheries in south
Wales. Journal of Environmental Management 25(2): 125-135.
Cullen, S.J.; Montagne, C.; Ferguson, H. 1991. Timber harvest trafficking
and soil compaction in western Montana. Soil Science Society of
America Journal 55(5): 1416-1421.
Cummins, K.W.; Sedell, J.R.; Swanson, F.J.; Minshall, G.W.; Fisher, S.G.;
Cushing, C.E.; Petersen, R.C.; Vannote, R.L. 1983. Organic matter
budgets for stream ecosystems: problems in their evaluation. In:
Barnes, J.R.; Minshall, G.W., eds. Stream ecology: application and
testing of general ecological theory. New York: Plenum Press; 299-353.
Cunjak, R.A. 1988. Behavior and microhabitat of young Atlantic salmon
(Salmo salar) during winter. Canadian Journal of Fisheries and Aquatic
Sciences 45(12): 2156-2160.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Dallal, G.E. 1988. Statistical microcomputing—like it is. American
Statistician 42(3): 212-216.
Danielsen, Karen C.; Halvorson, William L. 1991. Valley oak seedling
growth associated with selected grass species. Proceedings of the
symposium on oak woodlands and hardwood rangeland management;
1990 Oct. 31-Nov. 2; Davis, CA. Gen. Tech. Rep. PSW-126. Berkeley,
CA: Pacific Southwest Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 9-13.
Daubert, J.T.; Young, R.A. 1981. Recreational demands for maintaining
instream flows. American Journal of Agricultural Economics 63: 667-676.
Davidson, E.; Fox, M. 1974. Effects of off-road motorcycle activity on
Mojave Desert vegetation and soil. Madrono 22(8): 381-390.
Davis, E.A.; Ingebo, P.A. 1973. Picloram movement from the chaparral
watershed. Water Resources Research 9: 1304-1313.
Davis, J.C. 1976. Progress in sublethal effect studies with kraft pulpmill
effluent and salmonids. Journal of the Fisheries Research Board of
Canada 33(9): 2031-2035.
Day, N.F.; Megahan, W.F. 1975. Landslide occurrence on the Clearwater
National Forest. Geological Society of America Abstracts with Program.
7: 602-603.
Day, W.R. 1950. The soil conditions which determine windthrow in
forests. Forestry 23: 90-95.
Debano, L.F. 1981. Water-repellent soils: a state-of-the-art. Gen. Tech.
Rep. PSW-46. Berkeley, CA: Pacific Southwest Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture; 21 p.
DeByle, N.V.; Packer, P.E. 1972. Plant nutrient and soil losses in overland
flow from burned forest clearcuts. In: National symposium on
watersheds in transition. Fort Collins, CO: American Water Resources
Association and Colorado State University; 296-307.
Dedkov, A.P.; Moszherin, V.I.; Tchasovnikova, E.A. 1978. Field station
study of soil creep in central Volgaland. Zeitschrift für Geomorphologie
Neue Folge Supplementband 29: 111-116.
Dendy, F.E.; Champion, W.A. 1978. Sediment deposition in U.S.
reservoirs. Summary of data reported through 1975. Misc. Pub.
1362. Washington, DC: U.S. Department of Agriculture; 84 p.
Deverel, S.J.; Gallanthine, S.K. 1989. Relation of salinity and selenium
in shallow groundwater to hydrologic and geochemical processes,
western San Joaquin Valley, California. Journal of Hydrology 109(1/
2): 125-150.
Dickerson, B.P. 1976. Soil compaction after tree-length skidding in
northern Mississippi. Soil Science Society of America Journal 40:
965-966.
Dickert, T.G.; Tuttle, A.E. 1985. Cumulative impact assessment in
environmental planning. A coastal wetland watershed example.
Environmental Impact Assessment and Review 5: 37-64.
Dietrich, W.E.; Dunne, T. 1978. Sediment budget for a small catchment
in mountainous terrain. Zeitschrift für Geomorphologie Neue Folge
Supplementband 29: 191-206.
Dietrich, W.E.; Dunne, T.; Humphrey, N.F.; Reid, L.M. 1982. Construction
of sediment budgets for drainage basins. In: Proceedings of the
symposium of sediment budgets and routing in forest catchments. Gen.
Tech. Rep. PNW-141. Portland, OR: Pacific Northwest Forest and
Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 5-23.
Dietrich, W.E.; Kirchner, J.W.; Ikeda, H.; Iseya, F. 1989. Sediment supply
and the development of the coarse surface layer in gravel-bedded
rivers. Nature 340(6230): 215-217.
Dietterick, B.C.; Lynch, J.A. 1989. The cumulative hydrologic effects on
stormflows of successive clearcuts on a small headwater basin. In:
Headwaters hydrology. Bethesda, MD: American Water Resources
Association; 473-485.
Diseker, E.G.; Richardson, E.C. 1962. Erosion rates and control methods
on highway cuts. Transactions of the American Society of Agricultural
Engineers 5: 153-155.
Dissmeyer, G.E.; Foster, G.R. 1981. Estimating the cover-management
factor (C) in the universal soil loss equation for forest conditions.
Journal of Soil and Water Conservation 36(4): 235-240.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Doerr, T.B.; Redente, E.F.; Reeves, F.B. 1984. Effects of soil disturbance
on plant succession and levels of mycorrhizal fungi in a sagebrush-
grassland community. Journal of Range Management 37(2): 135-139.
Dolloff, C.A. 1986. Effects of stream cleaning on juvenile coho salmon
and Dolly Varden in southeast Alaska. Transactions of the American
Fisheries Society 115(5): 743-755.
Dolloff, C.A. 1987. Seasonal population characteristics and habitat use
by juvenile coho salmon in a small southeast Alaska stream.
Transactions of the American Fisheries Society 116(6): 829-838.
Dominy, C.L. 1973. Recent changes in Atlantic Salmon (Salmo salar)
runs in the light of environmental changes in the Saint John River,
New Brunswick, Canada. Biological Conservation 5(2): 105-113.
Dotzenko, A.D.; Papamichos, N.T.; Romine, D.S. 1967. Effect of
recreational use on soil and moisture conditions in Rocky Mountain
National Park. Journal of Soil and Water Conservation 22: 196-197.
Driver, N.E.; Troutman, B.M. 1989. Regression models for estimating
urban storm-runoff quality and quantity in the United States.
Journal of Hydrology 109(3/4): 221-236.
Duck, R.W. 1985. The effect of road construction on sediment deposition
in Loch Earn, Scotland. Earth Surface Processes and Landforms
10(4): 401-406.
Duda, A.M. 1986. Water quality impacts from sediment in the southeast.
In: Proceedings of the fourth federal interagency sedimentation
conference; 1986 March 24-27; Las Vegas, NV. Washington, DC:
Interagency Advisory Committee on Water Data, Subcommittee on
Sedimentation; 8-66 to 8-75.
Duffy, P.D.; McClurkin, D.C. 1974. Difficult eroded planting sites in
north Mississippi evaluated by discriminant analysis. Soil Science
Society of America Proceedings 38: 676-678.
Duffy, P.D.; Schreiber, J.D.; Ursic, S.J. 1986. Nutrient transport by
sediment from pine forests. In: Proceedings of the fourth federal
interagency sedimentation conference; 1986 March 24-27; Las Vegas,
NV. Washington, DC: Interagency Advisory Committee on Water
Data, Subcommittee on Sedimentation; 8-57 to 8-65.
Duijsings, J.J.H.M. 1987. A sediment budget for a forested catchment
in Luxembourg and its implications for channel development. Earth
Surface Processes and Landforms 12(2): 173-184.
Dunavin, L.S., Lutrick, M.C. 1983. Residual effects of liquid digested
sludge on the quality of broomsedge in a pine plantation. Journal of
Range Management 36(6): 773-774.
Duncan, S.H.; Bilby, R.E.; Ward, J.W.; Heffner, J.T. 1987. Transport of
road-surface sediment through ephemeral stream channels. Water
Resources Bulletin 23(1): 113-119.
Duncan, S.H.; Ward, J.W.; Anderson, R.J. 1987. A method for assessing
landslide potential as an aid in forest road placement. Northwest
Science 61(3): 152-159.
Dunne, T. 1984. The prediction of erosion in forests. In: O’Loughlin,
C.L.; Pearce, A.J., eds. Symposium on effects of forest land use on
erosion and slope stability; 1984 May 7-11; Honolulu, HI: East-West
Center, University of Hawaii; 3-11.
Dunne, T.; Leopold, L.B. 1978. Water in environmental planning. San
Francisco, CA: W.H. Freeman and Company; 818 p.
Durgin, P.B. 1985. Burning changes the erodibility of forest soils.
Journal of Soil and Water Conservation 40(3): 299-301.
Dyrness, C.T. 1965. Soil surface conditions following tractor and
high-lead logging in the Oregon Cascades. Journal of Forestry 63:
272-275.
Dyrness, C.T. 1967. Mass soil movements in the H.J. Andrews
Experimental Forest. Res. Paper PNW-42. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 12 p.
Dyrness, C.T. 1970. Stabilization of newly constructed road backslopes
by mulch and grass-legume treatments. Res. Paper PNW-123.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture.
101
Dyrness, C.T. 1972. Soil surface conditions following balloon logging.
Res. Note PNW-182. Portland, OR: Pacific Northwest Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture.
Dyrness, C.T. 1973. Early stages of plant succession following logging
and burning in the western Cascades of Oregon. Ecology 54: 57-69.
Dyrness, C.T. 1975. Grass-legume mixtures for erosion control along
forest roads in western Oregon. Journal of Soil and Water Conservation
30: 169-173.
Dyrness, C.T. 1976. Effect of wildfire on soil wettability in the High
Cascades of Oregon. Res. Paper PNW-202. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture.
Dzurisin, D. 1975. Channel responses to artificial stream capture,
Death Valley, California. Geology 3(6): 309-312.
Eckerberg, K. 1988. Clear felling and environmental protection: results
from an investigation in Swedish forests. Journal of Environmental
Management 27(3): 237-256.
Eckert, R.E., Jr.; Wood, M.K.; Blackburn, W.H.; Peterson, F.J. 1979.
Impacts of off-road vehicles on infiltration and sediment production
of two desert soils. Journal of Range Management 32(5): 394-397.
Edmonds, R.L. 1980. Litter decomposition and nutrient release in
Douglas-fir, red alder, western hemlock, and Pacific silver fir
ecosystems in western Washington. Canadian Journal of Forest
Research 10: 327-337.
Eiseley, L. 1946. The immense journey. New York: Random House.
Eissenstat, D.M.; Mitchell, J.E.; Pope, W.W. 1982. Trampling damage
by cattle on northern Idaho forest plantations. Journal of Range
Management 35(6): 715-716.
El-Hinnawi, E.E. 1981. The environmental impacts of production and
use of energy. Tycooly Press Ltd. for the United Nations Environmental
Program; 322 p.
El-Kadi, A.I. 1989. Watershed models and their applicability to conjunctive
use management. Water Resources Bulletin 25(1): 125-138.
Elliott, S.T. 1986. Reduction of a Dolly Varden population and
macrobenthos after removal of logging debris. Transactions of the
American Fisheries Society 115(3): 392-400.
Ellison, L.; Woolfolk, E.J. 1937. Effects of drought on vegetation near
Miles City, Montana. Ecology 18: 329-336.
Elsenrelch, S.J.; Metzer, N.A.; Urban, N.R.; Robbins, J.A. 1986. Response
of atmospheric lead to decreased use of lead in gasoline.
Environmental Science and Technology 20: 171-174.
Elwood, J.W.; Waters, T.F. 1969. Effects of floods on food consumption
and production rates of a stream brook trout population. Transactions
of the American Fisheries Society 98(2): 253-262.
Emmerich, W.E.; Woolhiser, D.A.; Shirley, E.D. 1989. Comparison of
lumped and distributed models for chemical transport by surface
runoff. Journal of Environmental Quality 18(1): 120-126.
Emmett, W.W. 1974. Channel aggradation in western U.S. as indicated
by observations at Vigil Network Sites. Zeitschrift für Geomorphologie
Neue Folge Supplementband 21: 52-62.
Englin, J. 1990. Backcountry hiking and optimal timber rotation. Journal
of Environmental Management 31(2): 97-106.
Erdman, J.A.; Ebens, R.J.; Case, A.A. 1978. Molybdenosis: a potential
problem in ruminants grazing on coal mine spoils. Journal of Range
Management 31: 34-36.
Erman, D.C.; Andrews, E.D.; Yoder-Williams, M. 1988. Effects of winter
floods on fishes in the Sierra Nevada. Canadian Journal of Fisheries
and Aquatic Sciences 45: 2195-2200.
Erman, D.C.; Newbold, J.D.; Roby, K.B. 1977. Evaluation of streamside
buffer strips for protecting aquatic organisms. Publ. no. 165. Davis,
CA: California Water Resources Center, University of California; 48 p.
Everett, R.L.; Sharrow, S.H. 1985. Understory response to tree harvesting
of singleleaf pinyon and Utah Juniper. Great Basin Naturalist 45(1):
105-112.
Farmer, E.E.; Fletcher, J.E. 1977. Highway erosion control systems: An
evaluation based on the Universal Soil Loss Equation. In: Soil
erosion: prediction and control. Proceedings of the national conference
102
on soil erosion; 1976 May 24-26; Purdue University, West Lafayette,
IN. Special Pub. 21. Washington, DC: Department of Agriculture, Soil
Conservation Service; 12-21.
Fausch, K.D. 1988. Tests of competition between native and introduced
salmonids in streams: what have we learned? Canadian Journal of
Fisheries and Aquatic Sciences 45: 2238-2246.
Feller, M.C. 1988. Relationships between fuel properties and
slashburning-induced nutrient losses. Forest Science 34(4): 998-1015.
Feller, M.C. 1989. Effects of forest herbicide applications on streamwater
chemistry in southwestern British Columbia. Water Resources Bulletin
25(3): 607-616.
Fenn, M.E.; Dunn, P.H. 1989. Litter decomposition across an air-pollution
gradient in the San Bernardino Mountains. Soil Science Society of
America Journal 53(5): 1560-1566.
Ffolliott, P.F.; Gottfried, G.J.; Baker, M.B., Jr. 1989. Water yield from
forest snowpack management: research findings in Arizona and
New Mexico. Water Resources Research 25(9): 1999-2007.
Fitzgerald, R.D.; Bailey, A.W. 1984. Control of aspen regrowth by
grazing with cattle. Journal of Range Management 37(2): 156-158.
Fleck, A.M.; Lacki, M.J.; Sutherland, J. 1988. Response by white birch
(Betula papyrifera) to road salt applications at Cascade Lakes, New
York. Journal of Environmental Management 27(4): 369-378.
Fogel, R.; Cromack, K., Jr. 1977. Effect of habitat and substrate
quality on Douglas-fir litter decomposition in western Oregon.
Canadian Journal of Botany 55: 1632-1640.
Ford, T.E.; Naiman, R.J. 1989. Groundwater-surface water relationships
in boreal forest watersheds—dissolved organic carbon and inorganic
nutrient dynamics. Canadian Journal of Fisheries and Aquatic Sciences
46(1): 41-49.
Forster, D.L.; Bardos, C.P.; Southgate, D.D. 1987. Soil erosion and water
treatment costs. Journal of Soil and Water Conservation 42(5): 349-351.
Fowler, W.B.; Anderson, T.D.; Helvey, J.D. 1988. Changes in water
quality and climate after forest harvest in central Washington
State. Res. Paper PNW-RP-388. Portland, OR: Pacific Northwest Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 12 p.
Fredriksen, R.L. 1963. A case study of a mud and rock slide on an
experimental watershed. Res. Note PNW-1. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 4 p.
Fredriksen, R.L. 1970. Erosion and sedimentation following road
construction and timber harvest on unstable soils in three small
western Oregon watersheds. Res. Paper PNW-104. Portland, OR:
Pacific Northwest Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 15 p.
Frissel, S.S., Jr.; Duncan, D.P. 1965. Campsite preference and
deterioration in the Quetico-Superior canoe country. Journal of
Forestry 63: 256-260.
Froehlich, H.A. 1979. Soil compaction from logging equipment: effects
on growth of young ponderosa pine. Journal of Soil and Water
Conservation 34: 276-278.
Froehlich, H.A.; Miles, D.W.R.; Robbins, R.W. 1985. Soil bulk density
recovery on compacted skid trails in central Idaho. Soil Science
Society of America Journal 49: 1015-1017.
Froehlich, W. 1991. Sediment production from unmetalled road surfaces.
In: Sediment and stream water quality in a changing environment:
trends and explanation. 203. International Association of Hydrological
Sciences; 21-30.
Fryer, J.H.; Ledig, F.T. 1972. Microevolution of the photosynthetic
temperature optimum in relation to the elevational complex gradient.
Canadian Journal of Botany 50: 1231-1235.
Furbish, D.J.; Rice, R.M. 1983. Predicting landslides related to clearcut
logging, northwestern California, USA. Mountain Research and
Development 3(3): 253-259.
Galbraith, A.F. 1975. Method for predicting increases in water yield
related to timber harvesting and site conditions. In: Water
management symposium; 1975 August; Logan, UT. American Society
of Civil Engineers; 169-184.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Garbrecht, J.; DeCoursey, D.G. 1986. A model relating water and sediment
yield to upstream agricultural practices. In: Proceedings of the
American Society of Civil Engineers water forum ’86 conference;
1986 Aug. 4-6; Long Beach, CA. American Society of Civil Engineers;
1759-1766.
Garrad, P.N.; Hey, R.D. 1988. River management to reduce turbidity in
navigable broadland rivers. Journal of Environmental Management
27(3): 273-288.
Gausen, D.; Moen, V. 1991. Large-scale escapes of farmed Atlantic
salmon (Salmo salar) into Norwegian rivers threaten natural
populations. Canadian Journal of Fisheries and Aquatic Sciences 48(3):
426-428.
Geist, J.M.; Hazard, J.W.; Seidel, K.W. 1989. Assessing physical conditions
of some Pacific Northwest volcanic ash soils after forest harvest.
Soil Science Society of America Journal 53(3): 946-950.
Gent, J.A., Jr.; Ballard, R. 1985. Impact of intensive forest management
practices on the bulk density of lower coastal plain and piedmont
soils. Southern Journal of Applied Forestry 9(1): 44-48.
Geppert, R.R.; Lorenz, C.W.; Larson, A.G. 1984. Cumulative effects of
forest practices on the environment. Prepared for the Washington
Forest Practices Board. Olympia, WA: Ecosystems, Inc.; 272 p.
Germanoski, D.; Ritter, D.F. 1987. Tributary entrenchment in response
to lowered base level below a dam. In: Erosion and sedimentation in
the Pacific Rim. Publ. no. 165. International Association of Hydrological
Sciences; 391-392.
Gholz, H.L.; Hawk, G.M.; Campbell, A.; Cromack, K., Jr. 1985. Early
revegetation recovery and element cycles on a clear-cut watershed
in western Oregon. Canadian Journal of Forest Research 15: 400-409.
Gibbens, R.P.; Beck, R.F. 1988. Changes in grass basal area and forb
densities over a 64-year period on grassland types of the Jornada
Experimental Range. Journal of Range Management 41(3): 186-192.
Gibbons, D.R.; Salo, E.O. 1973. An annotated bibliography of the
effects of logging on fish of the western U.S. and Canada. Gen.
Tech. Rep. PNW-10. Portland, OR: Pacific Northwest Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture.
Gibbs, M.J.; Hoffman, J.S. 1987. An approach for generating climate
change hypotheticals given limitations in current climate models.
In: The greenhouse effect, climate change, and U.S. forests. Washington,
DC: Conservation Foundation; 91-111.
Gibson, R.J.; Myers, R.A. 1988. Influence of seasonal river discharge on
survival of juvenile Atlantic salmon, Salmo salar. Canadian Journal
of Fisheries and Aquatic Sciences 45(2): 344-348.
Gifford, G.F.; Hawkins, R.H. 1978. Hydrologic impact of grazing on
infiltration: a critical review. Water Resources Research 14: 305-313.
Gilbert, G.K. 1917. Hydraulic-mining debris in the Sierra Nevada.
Prof. Paper 105. Washington, DC: Geological Survey, U.S. Department
of Interior; 154 p.
Gillen, R.L.; Krueger, W.C.; Miller, R.F. 1985. Cattle use of riparian
meadows in the Blue Mountains of northeastern Oregon. Journal of
Range Management 38(3): 205-209.
Gilley, J.E.; Finkner, S.C.; Varvel, G.E. 1987. Size distribution of sediment
as affected by surface residue and slope length. Transactions of the
American Society of Agricultural Engineers 30(5): 1419-1424.
Gilley, J.E.; Gee, G.W.; Bauer, A.; Willis, W.O.; Young, R.A. 1977.
Runoff and erosion characteristics of surface-mined sites in western
North Dakota. Transactions of the American Society of Agricultural
Engineers 20: 697-700, 704.
Gilliom, R.J.; Alexander, R.B.; Smith, R.A. 1985. Pesticides in the Nation’s
rivers, 1975-1980, and implications for future monitoring. Water
Supply Paper 2271. Washington, DC: Geological Survey, U.S.
Department of Interior; 26 p.
Gilliom, R.J.; Clifton, D.G. 1990. Organochlorine pesticide residues in
bed sediments of the San Joaquin River, California. Water Resources
Bulletin 26(1): 11-24.
Glancy, P.A.; Whitney, J.W. 1989. Recent hydrogeologic changes in
Lower Las Vegas Wash, Clark County, Nevada (abstract). Eos,
Transactions, American Geophysical Union 70(43): 1124.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Gold, A.J.; DeRagon, W.R.; Sullivan, W.M.; Lemunyon, J.L. 1990. Nitrate-
nitrogen losses to groundwater from rural and suburban land uses.
Journal of Soil and Water Conservation 45(2): 305-310.
Goldstein, R.A.; Mankin, J.B.; Luxmore, R.J. 1974. Documentation of
PROSPER: a model of atmosphere-soil-plant water flow. Publ. no.
579. Oak Ridge, TN: Oak Ridge National Laboratory, Environmental
Science Division; 75 p.
Gomez, B.; Church, M. 1989. An assessment of bedload sediment
transport formulae for gravel bed rivers. Water Resources Research
25(6): 1161-1186.
Gonsior, M.J.; Gardner, R.B. 1971. Investigation of slope failures in the
Idaho batholith. Res. Paper INT-97. Ogden, UT: Intermountain Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 34 p.
Govers, G. 1987. Initiation of motion in overland flow. Sedimentology
34(6): 1157-1164.
Govers, G.; Rauws, G. 1986. Transporting capacity of overland flow on
plane and on irregular beds. Earth Surface Processes and Landforms
11(5): 515-524.
Graf, W.L. 1975. The impact of suburbanization on fluvial
geomorphology. Water Resources Research 11: 690-692.
Graf, W.L. 1977. Network characteristics in suburbanizing streams.
Water Resources Research 13: 459-463.
Graf, W.L. 1978. Fluvial adjustments to the spread of tamarisk in the
Colorado Plateau region. Geological Society of America Bulletin 89:
1491-1501.
Graf, W.L. 1979. The development of montane arroyos and gullies.
Earth Surface Processes 4(1): 1-14.
Graf, W.L. 1980. The effect of dam closure on downstream rapids.
Water Resources Research 16(1): 129-136.
Graham, R.L.; Cromack, K., Jr. 1982. Mass, nutrient content, and decay
rate of dead boles in rain forests of Olympic National Park. Canadian
Journal of Forest Research 12: 511-521.
Grant, G. 1987. Assessing effects of peak flow increases on stream
channels: a rational approach. In: Proceedings of the California
watershed management conference; 1986 Nov. 18-20; West Sacramento,
CA. Rep. 11. Berkeley, CA: University of California, Wildland Resources
Center; 142-149.
Grant, G. 1988. The RAPID technique: a new method for evaluating
downstream effects of forest practices on riparian zones. Gen.
Tech. Rep. PNW-220. Portland, OR: Pacific Northwest Forest and
Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 36 p.
Grant, G.E.; Wolff, A.L. 1991. Long-term patterns of sediment transport
after timber harvest, western Cascade mountains, Oregon, USA.
In: Sediment and stream water quality in a changing environment:
trends and explanation. Publication 203. International Association of
Hydrological Sciences; 31-40.
Gray, D.H. 1970. Effects of forest clearcutting on the stability of natural
slopes. Bulletin of the Association of Engineering Geologists 7: 45-67.
Gray, D.H.; Megahan, W.F. 1981. Forest vegetation removal and slope
stability in the Idaho Batholith. Res. Paper INT-271. Ogden, UT:
Intermountain Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 23 pp.
Greacen, E.L.; Sands, R. 1980. Compaction of forest soils: a review.
Australian Journal of Soil Research 18: 163-189.
Green, W.D.; Stuart, W.B. 1985. Skidder and tire size effects on soil
compaction. Southern Journal of Applied Forestry 9(3): 154-157.
Gregory, K.J. 1977. Channel and network metamorphosis in northern
New South Wales. In: Gregory, K.J., ed. River channel changes. New
York: John Wiley and Sons.
Gregory, K.J.; Park, C. 1974. Adjustment of river channel capacity
downstream from a reservoir. Water Resources Research 10(4):
870-873.
Gresswell, S.; Heller, D.; Swanston, D.N. 1979. Mass movement response
to forest management in the central Oregon Coast Ranges. Res. Bull.
PNW-84. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 26 p.
103
Grier, C.C. 1975. Wildfire effects on nutrient distribution and leaching in
a coniferous ecosystem. Canadian Journal of Forest Research 5: 599-607.
Grimes, D.J. 1975. Release of sediment-bound fecal coliforms by
dredging. Applied Microbiology 29: 109-111.
Hack, J.T.; Goodlett, J.C. 1960. Geomorphology and forest ecology of a
mountain region in the central Appalachians. Prof. Paper 347.
Washington, DC: Geological Survey, U.S. Department of Interior.
Hagans, D.K.; Weaver, W.E. 1987. Magnitude, cause and basin response
to fluvial erosion, Redwood Creek basin, northern California. In:
Erosion and sedimentation in the Pacific Rim. Publ. 165. International
Association of Hydrological Sciences; 419-428.
Hall, W.S.; Dickson, K.L; Saleh, F.Y.; Rodgers, J.H., Jr.; Wilcox, D.;
Entazami, A. 1986. Effects of suspended solids on the acute toxicity
of zinc to Daphnia magna and Pimephales promelas. Water Resources
Bulletin 22(6): 913-920.
Halpern, C.B. 1988. Early successional pathways and the resistance
and resilience of forest communities. Ecology 69(6): 1703-1715.
Halpern, C.B. 1989. Early successional patterns of forest species:
interactions of life history traits and disturbance. Ecology 70(3):
704-720.
Hamburg, S.P.; Cogbill, C.V. 1988. Historical decline of red spruce
populations and climatic warming. Nature 331: 428-431.
Hamill, L. 1976. Analysis of Leopold’s quantitative comparisons of
landscape esthetics. Journal of Leisure Research 7(1): 16-28.
Hamill, L. 1986. The Chubb/Bauman method for estimating the
recreation potential of rivers: a critical review. Water Resources
Bulletin 22(4): 653-659.
Hamlett, J.M.; Baker, J.L.; Johnson, H.P. 1987. Size distribution of
sediment transported within and from small agricultural watersheds.
Transactions of the American Society of Agricultural Engineers 30(4):
998-1004.
Hammer, T.R. 1972. Stream channel enlargement due to urbanization.
Water Resources Research 8(6): 1530-1537.
Hansen, E.A.; Alexander, G.R. 1976. Effect of an artificially increased
sand bedload on stream morphology and its implications on fish
habitat. In: Proceedings of the third federal inter-agency sedimentation
conference; 1976 March 22-25; Denver, CO. Washington, DC: Water
Resources Council, Sedimentation Committee; 3-65 to 3-76.
Hansen, E.M. 1978. Incidence of Verticicladiella wageneri and Phellinus
weirii in Douglas-fir adjacent to and away from roads in western
Oregon. Plant Disease Reporter 62(2): 179-181.
Hansen, J.; Lacis, A.; Rind, D.; Russell, G.; Fung, I.; Lebedeff, S. 1987.
Evidence for future warming: how large and when? In: The
greenhouse effect, climate change, and U.S. forests. Washington, DC:
Conservation Foundation; 57-75.
Harden, D.R.; Janda, R.J.; Nolan, K.M. 1978. Mass movement and storms
in the drainage basin of Redwood Creek, Humboldt County,
California - a progress report. Open-file Report 78-486. Washington,
DC: Geological Survey, U.S. Department of Interior; 161 p.
Harmon, M.E.; Cromack, K., Jr.; Smith, B.G. 1987. Coarse woody debris
in mixed-conifer forests, Sequoia National Park, California. Canadian
Journal of Forest Research 17: 1265-1272.
Harmon, M.E.; Ferrell, W.K.; Franklin, J.F. 1990. Effects on carbon
storage of conversion of old-growth forests to young forests. Science
247: 699-702.
Harr, R.D. 1980. Streamflow after patch logging in small drainages
within the Bull Run Municipal Watershed, Oregon. Res. Paper
PNW-268. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 18 p.
Harr, R.D. 1986. Effects of clearcutting on rain-on-snow runoff in
western Oregon: A new look at old studies. Water Resources Research
22(7): 1095-1100.
Harr, R.D. 1987. Myths and misconceptions about forest hydrologic
systems and cumulative effects. In: Proceedings of the California
watershed management conference; 1986 November 18-20; West
Sacramento, CA. Publ. 11. Berkeley, CA: University of California,
Wildland Resources Center; 137-141.
104
Harr, R.D.; Fredriksen, R.L. 1988. Water quality after logging small
watersheds within the Bull Run watershed, Oregon. Water Resources
Bulletin 24(5): 1103-1112.
Harr, R.D.; Fredriksen, R.L.; Rothacher, J. 1979. Changes in streamflow
following timber harvest in southwest Oregon. Res. Paper PNW-
249. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 22 p.
Harr, R.D.; Harper, W.C.; Krygier, J.T.; Hsieh, F.S. 1975. Changes in
storm hydrographs after roadbuilding and clearcutting in the Oregon
Coast Range. Water Resources Research 11: 436-444.
Harr, R.D.; Levno, A.; Mersereau, R. 1982. Streamflow changes after
logging 130-year-old Douglas fir in two small watersheds. Water
Resources Research 18(3): 637-644.
Harr, R.D.; McCorison, F.M. 1979. Initial effects of clearcut logging on
size and timing of peak flows in a small watershed in western
Oregon. Water Resources Research 15: 90-94.
Harrington, T.C.; Reinhart, C.; Thornburgh, D.A.; Cobb, F.W., Jr. 1983.
Association of black-stain root disease with precommercial thinning
of Douglas-fir. Forest Science 29(1): 12-14.
Harris, A.R.; Urie, D.H. 1983. Changes in a sandy forest soil under
northern hardwoods after 5 years of sewage effluent irrigation.
Soil Science Society of America Journal 47(4): 800-805.
Harris, A.S. 1966. Effects of slash burning on conifer regeneration in
southeast Alaska. Res. Note NOR-18. St. Paul, MN: Northern Forest
Experiment Station, Forest Service, U.S. Department of Agriculture; 6 p.
Harris, D.D. 1977. Hydrologic changes after logging in two small Oregon
coastal watersheds. Survey Water Supply Paper 2037. Washington,
DC: Geological Survey, U.S. Department of Interior.
Harris, R.R.; Fox, C.A.; Risser, R. 1987. Impacts of hydroelectric
development on riparian vegetation in the Sierra Nevada Region,
California, USA. Environmental Management 11(4): 519-527.
Hart, G.E.; Loomis, S.A. 1982. Erosion from mountain snowpack melt.
Journal of Soil and Water Conservation 37(1): 55-57.
Hartman, G.F.; Brown, T.G. 1987. Use of small, temporary, floodplain
tributaries by juvenile salmonids in a west coast rain-forest drainage
basin, Carnation Creek, British Columbia. Canadian Journal of
Fisheries and Aquatic Sciences 44(2): 262-270.
Harvey, M.D.; Schumm, S.A. 1987. Response of Dry Creek, California,
to land use change, gravel mining and dam closure. In: Erosion and
sedimentation in the Pacific Rim. Publication 165. International
Association of Hydrological Sciences; 451-460.
Haskins, D.M. 1987. A management model for evaluating cumulative
watershed effects. In: Proceedings of the California watershed
management conference; 1986 Nov. 18-20; West Sacramento, CA.
Publ. 11. Berkeley, CA: University of California, Wildland Resources
Center; 125-130.
Hatchell, G.E.; Ralston, C.W.; Foil, R.R. 1970. Soil disturbances in
logging, effects on soil characteristics and growth of loblolly pine in
the Atlantic Coastal Plain. Journal of Forestry 68: 772-775.
Haupt, H.F. 1959. Road and slope characteristics affecting sediment
movement from logging roads. Journal of Forestry 57: 329-332.
Hawley, J.G.; Dymond, J.R. 1988. How much do trees reduce landsliding?
Journal of Soil and Water Conservation 43(6): 495-498.
Haws, E.T.; Reilly, N. 1981. Dams, natural and induced earthquakes
and the environment. In: Dams and earthquake. London: Thomas
Telford Limited; 253-260.
Heady, H.F. 1961. Continuous vs. specialized grazing systems: a review
and application to the California annual type. Journal of Range
Management 14: 182-193.
Heede, B.H. 1971. Characteristics and processes of soil piping in gullies.
Res. Paper RM-68. Fort Collins, CO: Rocky Mountain Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture; 15 p.
Heede, B.H. 1972. Flow and channel characteristics of two high mountain
streams. Res. Paper RM-96. Fort Collins, CO: Rocky Mountain Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Heede, B.H. 1984. Sediment source areas related to timber harvest on
selected Arizona watersheds. In: O’Loughlin, C.L.; Pearce, A.J., eds.
Proceedings of a symposium on effects of forest land use on erosion
and slope stability; 1984 May 7-11; Honolulu, HI. Honolulu, HI:
University of Hawaii, East-West Center; 123-130.
Heede, B.H.; Harvey, M.D.; Laird, J.R. 1988. Sediment delivery linkages
in a chaparral watershed following a wildfire. Environmental
Management 12(3): 349-358.
Heede, B.H.; King, R.M. 1990. State-of-the-art timber harvest in an
Arizona mixed conifer forest has minimal effect on overland flow
and erosion. Hydrological Sciences Journal 35(6): 623-636.
Heggenes, J. 1988. Substrate preferences of brown trout fry (Salmo
trutta) in artificial stream channels. Canadian Journal of Fisheries
and Aquatic Sciences 45(10): 1801-1806.
Helvey, J.D.; Fowler, W.B. 1979. Grass seeding and soil erosion in a
steep, logged area in northeastern Oregon. Res. Note PNW-343.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 11 p.
Helvey, J.D.; Kochenderfer, J.N. 1987. Effects of limestone gravel on
selected chemical properties of road runoff and streamflow. Northern
Journal of Applied Forestry 4(1): 23-25.
Helvey, J.D.; Tiedemann, A.R.; Anderson, T.D. 1985. Plant nutrient
losses by soil erosion and mass movement after wildfire. Journal of
Soil and Water Conservation 40(1): 168-173.
Hemstrom, M.A.; Franklin, J.F. 1982. Fire and other disturbances of the
forests in Mount Rainier National Park. Quaternary Research 18:
32-51.
Herbel, C.H. 1985. Vegetation changes on arid rangelands of the
southwest. Rangelands 7(1): 19-21.
Hess, S. 1984. Timber harvesting and flooding. Journal of Soil and
Water Conservation 39:115-117.
Hewlett, J.F.; Helvey, J.D. 1970. Effects of forest clear-felling on the
storm hydrograph. Water Resources Research 6(3): 768-782.
Heywood, J.L. 1987. Experience preferences of participants in different
types of river recreation groups. Journal of Leisure Research 19(1):
1-12.
Hicks, B.J.; Beschta, R.L.; Harr, R.D. 1991. Long-term changes in
streamflow following logging in western Oregon and associated
fisheries implications. Water Resources Bulletin 27(2): 217-226.
Higgins, D.A.; Maloney, S.B; Tiedemann, A.R.; Quigley, T.M. 1989.
Storm runoff characteristics of grazed watersheds in eastern Oregon.
Water Resources Bulletin 25(1): 87-100.
Higgins, K.F. 1984. Lightning fires in North Dakota grasslands and in
pine-savanna lands of South Dakota and Montana. Journal of Range
Management 37(2): 100-103.
Hill, R.L. 1990. Long-term conventional and no-tillage effects on selected
soil physical properties. Soil Science Society of America Journal
54(1): 161-166.
Hillman, G.R.; Verschuren, J.P. 1988. Simulation of the effects of forest
cover, and its removal, on subsurface water. Water Resources Research
24(2): 305-314.
Hillman, T.W.; Griffith, J.S.; Platts, W.S. 1987. Summer and winter
habitat selection by juvenile chinook salmon in a highly sedimented
Idaho stream. Transactions of the American Fisheries Society 116:
185-195.
Hinrichsen, D. 1986. Multiple pollutants and forest decline. Ambio
15(5): 258-265.
Hirschi, M.C.; Barfield, B.J. 1988a. KYERMO - A physically based
research erosion model Part I. Model development. Transactions of
the American Society of Agricultural Engineers 31(3): 804-813.
Hirschi, M.C.; Barfield, B.J. 1988b. KYERMO - A physically based
research erosion model Part II. Model sensitivity analysis and
testing. Transactions of the American Society of Agricultural Engineers
31(3): 814-820.
Hoar, S.K.; Blair, A.; Holmes, F.F.; Boyson, C.D.; Robel, R.J.; Hoover,
R.; Fraumeni, J.F., Jr. 1986. Agricultural herbicide use and risk of
lymphoma and soft-tissue sarcoma. Journal of the American Medical
Association 256(9): 1141-1147.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Hoffman, M.S., ed. 1990. The world almanac and book of facts 1991.
New York: Pharos Books; 960 p.
Hofstra, G.; Hall, R. 1971. Injury on roadside trees: leaf injury on pine
and white cedar in relation to foliar levels of sodium chloride.
Canadian Journal of Botany 49: 613-622.
Hofstra, G.; Smith, D.W. 1984. The effects of road de-icing salt on the
levels of ions in roadside soils in southern Ontario. Journal of
Environmental Management 19: 261-271.
Hogan, D.L. 1987. The influence of large organic debris on channel
recovery in the Queen Charlotte Islands, British Columbia, Canada.
In: Erosion and sedimentation in the Pacific Rim. Publ. 165. International
Association of Hydrological Sciences; 343-353.
Holberger, R.L.; Truett, J.B. 1976. Sediment yield from construction
sites. In: Proceedings of the third federal inter-agency sedimentation
conference; 1976 March 22-25; Denver, CO. Washington, DC: Water
Resources Council, Sedimentation Committee; I-47 to I-58.
Holechek, J.L.; Depuit, E.J.; Coenenberg, J.; Valdez, R. 1982. Long-term
plant establishment on mined lands in southeastern Montana. Journal
of Range Management 35(4): 522-525.
Hollis, G.E. 1975. The effect of urbanization on floods of different
recurrence intervals. Water Resources Research 11: 431-435.
Holtby, L.B. 1988. Effects of logging on stream temperatures in Carnation
Creek, British Columbia, and associated impacts on the coho salmon
(Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic
Sciences 45(3): 502-515.
Hooke, J.M. 1979. An analysis of the processes of bank erosion. Journal
of Hydrology 42: 39-62.
Hooke, J.M. 1980. Magnitude and distribution of rates of river bank
erosion. Earth Surface Processes 5(2): 143-157.
Hooven, E.F; Black, H.C. 1978. Prescribed burning aids reforestation
of Oregon Coast Range brushlands. Res. Paper 38. Corvallis, OR:
Oregon State University Forest Research Laboratory.
Hornbeck, J.W. 1973. Stormflow from hardwood forested and cleared
watersheds in New Hampshire. Water Resources Research 9(2):
346-354.
Hornbeck, J.W.; Martin, C.W.; Pierce, R.S.; Borman, F.H.; Likens, G.E.;
Eaton, J.S. 1986. Clearcutting northern hardwoods: effects on
hydrologic and nutrient ion budgets. Forest Science 32(3): 667-686.
Hornbeck, J.W.; Pierce, R.S.; Federer, C.A. 1970. Streamflow changes
after forest clearing in New England. Water Resources Research
6(4): 1124-1132.
Hornbeck, J.W.; Reinhart, K.G. 1964. Water quality and soil erosion as
affected by logging in steep terrain. Journal of Soil and Water
Conservation 19:23-27.
Hubbs, C. 1972. Some thermal consequences of environmental
manipulations of water. Biological Conservation 4(3): 185-188.
Hughs, R.M.; Gammon, J.R. 1987. Longitudinal changes in fish
assemblages and water quality in the Willamette River, Oregon.
Transactions of the American Fisheries Society 116(2): 196-209.
Huryn, A.D.; Wallace, J.B. 1987. Local geomorphology as a determinant
of macrofaunal production in a mountain stream. Ecology 68(6):
1932-1942.
Imhoff, K.R.; Koppe, P.; Dietz, F. 1980. Heavy metals in the Ruhr River
and their budget in the catchment area. Progress in Water Technology
12: 735-749.
Jackson, W.L.; Van Haveren, B.P. 1987. Predicting channel responses to
changing flow regimes: Beaver Creek, Alaska. In: Erosion and
sedimentation in the Pacific Rim. Publ. 165. International Association
of Hydrological Sciences; 393-394.
Jacob, G.R.; Schreyer, R. 1980. Conflict in outdoor recreation: a
theoretical perspective. Journal of Leisure Research 12(4): 368-380.
Jacobsen, T.; Adams, R.M. 1958. Salt and silt in ancient Mesopotamian
agriculture. Science 128: 1251-1258.
Jawson, M.D; Elliott, L.F.; Saxton, K.E.; Fortier, D.H. 1982. The effect of
cattle grazing on indicator bacteria in runoff from a Pacific Northwest
watershed. Journal of Environmental Quality 11(4): 621-627.
Jobson, H.E.; Carey, W.P. 1989. Interaction of fine sediment with alluvial
streambeds. Water Resources Research 25(1): 135-140.
105
Johansen, J.R.; Javakul, A.; Rushforth, S.R. 1982. Effects of burning on
the algal communities of a high desert soil near Wallsburg, Utah.
Journal of Range Management 35(5): 598-600.
Johnson, C.W.; Smith, J.P. 1983. Soil loss caused by off-road vehicle use
on steep slopes. Transactions of the American Society of Agricultural
Engineers 26(2): 402-405, 411.
Johnson, D.W.; Turner, J.; Kelley, J.M. 1982. The effects of acid rain on
forest nutrient status. Water Resources Research 18(3): 449-461.
Johnson, M.G.; Beschta, R.L. 1980. Logging, infiltration capacity, and
surface erodibility in western Oregon. Journal of Forestry 78(6):
334-337.
Johnson, N.S.; Adams, R.M. 1988. Benefits of increased streamflow: the
case of the John Day River steelhead fishery. Water Resources
Research 24(11): 1839-1846.
Johnson, R.C. 1988. Changes in the sediment output of two upland
drainage basins during forestry land use changes. In: Sediment
budgets. Publ no. 174. International Association of Hydrological
Sciences; 463-471.
Johnson, S.R.; Gary, H.L.; Ponce, S.L. 1978. Range cattle impacts on
stream water quality in Colorado Front Range. Res. Note RM-359.
Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture.
Johnston, C.A.; Bubenzer, G.D.; Lee, G.B.; Madison, F.W.; McHenry, J.R.
1984. Nutrient trapping by sediment deposition in a seasonally flooded
lakeside wetland. Journal of Environmental Quality 13: 283-290.
Johnston, C.A.; Detenbeck, N.E.; Bonde, J.P.; Niemi, G.J. 1988. Geographic
information systems for cumulative impact assessment.
Photogrammetric Engineering and Remote Sensing 54(11): 1609-1615.
Johnstone, I.M.; Coffey, B.T.; Howard-Williams, C. 1985. The role of
recreational boat traffic in interlake dispersal of macrophytes: a
New Zealand case study. Journal of Environmental Management
20(3): 263-279.
Jones, R.C.; Clark, C.C. 1987. Impact of watershed urbanization on
stream insect communities. Water Resources Bulletin 23(6): 1047-
1056.
Jury, W.A.; Gruber, J. 1989. A stochastic analysis of the influence of soil
and climatic variability on the estimate of pesticide groundwater
pollution potential. Water Resources Research 25(12): 2465-2474.
Kahan, A.M. 1972. Status of cloud seeding - 1972. In: National symposium
on watersheds in transition. American Water Resources Association
and Colorado State University, Fort Collins, CO; 206-211.
Kalisz, P.J. 1986. Soil properties of steep Appalachian old fields. Ecology
67(4): 1011-1023.
Kauffman, J.B.; Krueger, W.C. 1984. Livestock impacts on riparian
ecosystems and streamside management implications...a review.
Journal of Range Management 37(5): 430-438.
Kauffman, J.B.; Krueger, W.C.; Vavra, M. 1983a. Impacts of cattle on
streambanks in northeastern Oregon. Journal of Range Management
36(6): 683-684.
Kauffman, J.B.; Krueger, W.C.; Vavra, M. 1983b. Effects of late season
cattle grazing on riparian plant communities. Journal of Range
Management 36(6): 685-691.
Kauffman, J.B.; Martin, R.E. 1989. Fire behavior, fuel consumption,
and forest-floor changes following prescribed understory fires in
Sierra Nevada mixed conifer forests. Canadian Journal of Forest
Research 19(4): 455-462.
Keller, E.A.; Swanson, F.J. 1978. Effects of large organic material on
channel form and fluvial processes. Earth Surface Processes 4: 361-380.
Keller, E.A.; Tally, T. 1979. Effects of large organic debris on channel
form and fluvial processes in the coastal redwood environment. In:
Rhodes, D.D.; Williams, G.P., eds. Adjustments of the fluvial system.
Proceedings of the 10th annual geomorphology symposium, State
University of New York, Binghampton. Dubuque, IA: Kendall/Hunt
Publication Co.; 167-197.
Kelsey, H.M. 1980. A sediment budget and an analysis of geomorphic
processes in the Van Duzen River basin, north coastal California, 1941-
1975. Geological Society of America Bulletin Part II 91: 1119-1216.
106
Kelsey, H.M.; Lamberson, R.; Madej, M.A. 1987. Stochastic model for
the long-term transport of stored sediment in a river channel.
Water Resources Research 23(9): 1738-1750.
Kenimer, A.L.; Mostaghimi, S.; Dillaha, T.A.; Shanholtz, V.O. 1989.
PLIERS: Pesticide losses in erosion and runoff simulator.
Transactions of the American Society of Agricultural Engineers 32(1):
127-136.
Keppeler, E.T.; Ziemer, R.R. 1990. Logging effects on streamflow: water
yield and summer low flows at Caspar Creek in northwestern
California. Water Resources Research 26(7): 1669-1680.
Khanbilvardi, R.M.; Rogowski, A.S. 1984. Quantitative evaluation of
sediment delivery ratios. Water Resources Bulletin 20(6): 865-874.
Kidd, W.J. 1963. Soil erosion control structures on skid trails. Res.
Paper INT-1. Ogden, UT: Intermountain Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 8 p.
Kilinc, M.; Richardson, E.V. 1973. Mechanics of soil erosion from
overland flow generated by simulated rainfall. Hydrol. Paper 63.
Fort Collins, CO: Colorado State University; 54 p.
Kimmins, J.P. 1975. Review of the ecological effects of herbicide usage
in forestry. Information Report BC-X-139. Victoria, BC: Canadian
Forest Service, Pacific Forestry Centre; 44 p.
Kimmins, J.P. 1977. Evaluation of the consequences for future tree
productivity of the loss of nutrients in whole-tree harvesting. Forest
Ecology and Management 1(2): 169-183.
King, J.G. 1989. Streamflow responses to road building and harvesting:
a comparison with the equivalent clearcut area procedure. Res.
Paper INT-401. Ogden, UT: Intermountain Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 13 p.
King, J.G.; Tennyson, L.C. 1984. Alteration of streamflow characteristics
following road construction in north central Idaho. Water Resources
Research 20(8): 1159-1163.
King, N.J. 1961. An example of channel aggradation induced by flood
control. Prof. Paper 424B. Washington, DC: Geological Survey, U.S.
Department of Interior; 29-32.
Kingery, J.L. 1983. Research response to forestland grazing needs on
public lands of the northern intermountain region. In: Roche, B.F.,
ed. Forestland grazing. Proceedings of a symposium held 1983 Feb 23-
25; Spokane, WA; 85-87.
Kleerekoper, H. 1976. Effects of sublethal concentrations of pollutants
on the behavior of fish. Journal of the Fisheries Research Board of
Canada 33:2036-2039.
Klein, R.; Sonnevil, R.; Short, D. 1987. Effects of woody debris removal
on sediment storage in a northwest California stream. In: Erosion
and sedimentation in the Pacific Rim. Publ. 165. International Association
of Hydrological Sciences; 403-404.
Klein, R.M.; Perkins, T.D. 1987. Cascades of causes and effects of forest
decline. Ambio 16(2-3): 86-93.
Kleiner, E.F. 1983. Successional trends in an ungrazed, arid grassland
over a decade. Journal of Range Management 36(1): 114-118.
Klock, G.O. 1975. Impact of five postfire salvage logging systems on soil
and vegetation. Journal of Soil and Water Conservation 30(2): 78-81.
Klock, G.O. 1979. Some effects of soil erosion on forest and range soil
productivity. In: Agronomy Abstracts 1979 Annual Meeting; 206.
Klock, G.O. 1985. Modeling the cumulative effects of forest practices
on downstream aquatic ecosystems. Journal of Soil and Water
Conservation 40(2): 237-241.
Klock, G.O.; Helvey, J.P. 1976. Debris flows following wildfire in north
central Washington. In: Proceedings of the third federal inter-agency
sedimentation conference; 1976 March 22-25; Denver, CO. Washington,
DC: Water Resources Council, Sedimentation Committee; 1-91 to 1-98.
Knight, H. 1966. Loss of nitrogen from the forest floor by burning.
Forest Chronicles 42:149-152.
Knighton, A.D. 1988. The impact of the Parangana Dam on the River
Mersey, Tasmania. Geomorphology 1(3): 221-238.
Knisel, W.C., ed. 1980. CREAMS: A field scale model for chemicals,
runoff, and erosion from agricultural management systems.
Conservation Res. Report 26. Washington, DC: U.S. Department of
Agriculture; 643 p.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Knopp, T.B. 1972. Environmental determinants of recreation behavior.
Journal of Leisure Research 4(2): 129-138.
Knox, J.C.; Bartlein, P.J.; Hirschboeck, K.K.; Muckenhirn, R.J. 1975. The
response of floods and sediment yields to climatic variation and
land use in the upper Mississippi Valley. Publ. 52. Madison, WI:
University of Wisconsin, Institute of Environmental Studies; 76 p.
Kochenderfer, J.N.; Helvey, J.D. 1987. Using gravel to reduce soil losses
from minimum-standard forest roads. Journal of Soil and Water
Conservation 42(1): 46-50.
Kondolf, G.M.; Curry, R.R. 1986. Channel erosion along the Carmel
River, Monterey County, California. Earth Surface Processes and
Landforms 11(3): 307-320.
Kosco, H.; Bartolome, J.W. 1983. Effects of cattle and deer on
regenerating mixed conifer clearcuts. Journal of Range Management
36(2): 265-268.
Kozel, S.J.; Hubert, W.A. 1989. Factors influencing the abundance of
brook trout (Salvelinus fontinalis) in forested mountain streams.
Journal of Freshwater Ecology 5(1): 113-122.
Kozel, S.J.; Hubert, W.A.; Parsons, M.G. 1989. Habitat features and
trout abundance relative to gradient in some Wyoming streams.
Northwest Science 63(4): 175-182.
Kraemer, J.F.; Hermann, R.K. 1979. Broadcast burning: 25-year effects
on forest soils in the western flanks of the Cascade Mountains.
Forest Science 25(3): 427-439.
Kraft, J.C.; Rapp, G., Jr.; Aschenbrenner, S.E. 1975. Late Holocene
paleogeography of the coastal plain of the Gulf of Messenia, Greece,
and its relationships to archaeological settings and coastal change.
Geological Society of America Bulletin 86:1191-1208.
Krammes, J.S.; Burns, D.M. 1973. Road construction on Casper Creek
watersheds. Ten year report on impact. Res. Paper PSW-93. Berkeley,
CA: Pacific Southwest Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture.
Krebs, C.J. 1978. Ecology. New York: Harper and Row; 678 p.
LaHusen, R.G. 1984. Characteristics of management-related debris
flows, northwestern California. In: O’Loughlin, C.L.; Pearce, A.J.,
eds. In: Proceedings of a symposium on effects of forest land use on
erosion and slope stability; 1984 May 7-11; Honolulu, HI. Honolulu,
HI: University of Hawaii, East-West Center; 139-146.
Lane, L.J.; Schertz, D.L.; Alberts, E.E.; Laflen, J.M.; Lopes, V.L. 1988.
The U.S. national project to develop improved erosion prediction
technology to replace the USLE. In: Sediment budgets. Publ. no. 174.
International Association of Hydrological Sciences; 473-482.
LaPerriere, J.D.; Wagener, S.M.; Bjerklie, D.M. 1985. Gold-mining effects
on heavy metals in streams, Circle Quadrangle, Alaska. Water
Resources Bulletin 21: 245-252.
Larson, K.R.; Sidle, R.C. 1980. Erosion and sedimentation data catalog
of the Pacific Northwest. R6-WM-050-1981. Pacific Northwest Region.
Portland, OR: Forest Service, U.S. Department of Agriculture.
Larsson, U.; Elmgren, R.; Wulff, F. 1985. Eutrophication and the Baltic
Sea: causes and consequences. Ambio 14(1): 9-14.
Lawrence, G.B.; Driscoll, C.T. 1988. Aluminum chemistry downstream
of a whole-tree-harvested watershed. Environmental Science and
Technology 22(11): 1293-1298.
Leaf, C.F. 1974. A model for predicting erosion and sediment yield
from secondary forest road construction. Res. Note RM-274. Fort
Collins, CO: Rocky Mountain Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture.
Leaf, C.F.; Brink, G.E. 1973a. Computer simulation of snowmelt within
a Colorado subalpine watershed. Res. Paper RM-99. Fort Collins,
CO: Rocky Mountain Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 22 p.
Leaf, C.F., Brink, G. E. 1973b. Hydrologic simulation model of Colorado
subalpine forest. Res. Paper RM-107. Fort Collins, CO: Rocky Mountain
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture; 23 p.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Leavesley, G.H.; Lumb, A.M.; Saindon, L.G. 1987. A microcomputer-
based watershed-modeling and data-management system. In:
Proceedings of the western snow conference, Vancouver, BC; 108-117.
Ledig, F.T. 1986. Heterozygosity, heterosis, and fitness in outbreeding
plants. In: Soule, M.E., ed. Conservation Biology. Sinaur Press; 77-104.
Lehre, A.K. 1982. Sediment budget of a small mountain catchment in
north-central California. Proceedings of the symposium on sediment
budgets and routing in forest catchments. Gen. Tech. Rep. PNW-141.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture.
Lehre, A.K.; Collins, B.D.; Dunne, T. 1983. Post-eruption sediment
budget for the North Fork Toutle River Drainage, June 1980-June
1981. Zeitschrift für Geomorphologie Neue Folge Supplementband 46:
143-163.
Lein, J.K. 1989. The assessment of land use/land cover impact on local
climate. Journal of Environmental Management 28(2): 157-174.
Leininger, W.C.; Sharrow, S.H. 1987. Seasonal diets of herded sheep
grazing Douglas-fir plantations. Journal of Range Management 40(6):
551-554.
Leonard, R.E.; McMahon, J.L.; Kehoe, K.M. 1985. Hiker trampling
impacts on eastern forests. Res. Paper NE-555. Radnor, PA:
Northeastern Forest Experiment Station, Forest Service, U.S. Department
of Agriculture.
Leopold, A. 1949. A Sand County almanac, and sketches here and
there. New York: Oxford University Press; 226 p.
Leopold, L.B. 1969. Quantitative comparison of some aesthetic factors
among rivers. Circular 620. Washington, DC: Geological Survey,
U.S. Department of Interior; 16 p.
Leopold, L.B. 1973. River channel changes with time: an example.
Geological Society of America Bulletin 84:1845-1860.
Leopold, L.B.; Emmett, W.W; Myrick, R.M. 1966. Channel and hillslope
processes in a semiarid area, New Mexico. Prof. Paper 352G.
Washington, DC: Geological Survey, U.S. Department of Interior;
193-253.
Lettenmaier, D.P.; Gan, T.Y. 1990. Hydrologic sensitivities of the
Sacramento - San Joaquin River Basin, California, to global
warming. Water Resources Research 26(1): 69-86.
Lewin, J.; Davies, B.E.; Wolfenden, P.J. 1977. Interactions between
channel change and historic mining sediments. In: Gregory, K.J.,
ed. River channel changes. New York: John Wiley and Sons; 357-367.
Lewis, C.E; Tanner, G.W.; Terry, W.S. 1988. Plant responses to pine
management and deferred-rotation grazing in north Florida. Journal
of Range Management 41(6): 460-465.
Lewis, D.C.; Burgy, R.H. 1964. Hydrologic balance from an experimental
watershed. Journal of Hydrology 2: 197-212.
Lewis, H.T. 1973. Patterns of indian burning in California: ecology
and ethnohistory. Anthropological Papers 1. Ramona, CA: Ballena
Press; 101.
Lewis, J.; Rice, R.M. 1989. Critical sites erosion study: Volume II. Site
conditions related to erosion on private timberlands in northern
California. Final report to California Dept. Forestry. Berkeley, CA:
Pacific Southwest Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 95 p.
Lewis, J.; Rice, R.M. 1990. Estimating erosion risk on forest lands
using improved methods of discriminant analysis. Water Resources
Research 26(8): 1721-1734.
Li, C.Y.; Maser, C.; Maser, Z.; Caldwell, B.A. 1986. Role of three
rodents in forest nitrogen fixation in western Oregon: another
aspect of mammal-mycorrhizal fungus-tree mutualism. Great Basin
Naturalist 46(3): 411-414.
Liddle, M.J.; Scorgie, H.R.A. 1980. The effect of recreation on freshwater
plants and animals: a review. Biological Conservation 17: 183-206.
Lindsey, D.L.; Cress, W.A.; Aldon, E.F. 1977. The effects of
endomycorrhizal on growth of rabbitbrush, four-wing saltbush,
and corn in coal mine spoil material. Res. Note RM-343. Fort Collins,
CO: Rocky Mountain Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture.
107
Lisle, T.E. 1987. Overview: channel morphology and sediment transport
in steepland streams. In: Erosion and sedimentation in the Pacific
Rim. Publ. no 165. International Association of Hydrological Sciences;
287-297.
Lisle, T.E. 1989. Sediment transport and resulting deposition in spawning
gravels, north coastal California. Water Resources Research 25(6):
1303-1320.
Little, S.N.; Ohmann, J.L. 1988. Estimating nitrogen lost from forest
floor during prescribed fires in Douglas fir/western hemlock
clearcuts. Forest Science 34(1): 152-164.
Little, S.N.; Waddell, D.R. 1987. Highly stocked coniferous stands on
the Olympic Peninsula: chemical composition and implications for
harvest strategy. Res. Paper PNW-384. Portland, OR: Pacific Northwest
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture; 29 p.
Lockaby, B.G.; Vidrine, C.G. 1984. Effect of logging equipment traffic
on soil density and growth and survival of young loblolly pine.
Southern Journal of Applied Forestry 8:109-112.
Lockery, A.R.; Gavrailoff, T.; Hatcher, D. 1983. Lead levels in snow
dumping sites along rivers in downtown Winnipeg, Manitoba,
Canada. Journal of Environmental Management 17(2): 185-190.
Loope, W.L.; Gifford, G.F. 1972. Influence of a soil micro-floral crust
on select property of soils under pinyon-juniper in southeastern
Utah. Journal of Soil and Water Conservation 27:164-167.
Lorber, M.N.; Mulkey, L.A. 1982. An evaluation of three pesticide runoff
loading models. Journal of Environmental Quality 11: 519-529.
Lorimer, C.G. 1989. Relative effects of small and large disturbances on
temperate hardwood forest structure. Ecology 70(3): 565-567.
Loring, M.W.; Workman, J.P. 1987. The relationship between land
ownership and range condition in Rich County, Utah. Journal of
Range Management 40(4): 290-293.
Lorz, H.W.; Glenn, S.W.; Williams, R.H.; Kunkel, C.M.; Norris, L.A.;
Loper, B.R. 1979. Effects of selected herbicides on smolting of coho
salmon. EPA-600/3-79-071. Corvallis, OR: U.S. Environmental
Protection Agency, Environmental Research Laboratory; 103 p.
Lu, J.Y.; Cassol, E.A.; Foster, G.R.; Neibling, W.H. 1988. Selective
transport and deposition of sediment particles in shallow flow.
Transactions of the American Society of Agricultural Engineers 31(4):
1141-1147.
Luedtke, R.J.; Brusven, M.A. 1976. Effects of sand sedimentation on
colonization of stream insects. Journal of the Fisheries Research
Board of Canada 33: 1881-1886.
Lusby, G.C. 1970. Hydrologic and biotic effects of grazing versus nongrazing
near Grand Junction, Colorado. Prof. Paper 700B. Washington, DC:
Geology Survey, U.S. Department of Interior; 232-236.
Lyle, W.M.; Smerdon, E.T. 1965. Relation of compaction and other soil
properties of erosion resistance of soils. Transactions of the American
Society of Agricultural Engineers 8(3): 419-422.
Lyons, J.K.; Beschta, R.L. 1983. Land use, floods, and channel changes:
Upper Middle Fork Willamette River, Oregon (1936-1980). Water
Resources Research 19(2): 463-471.
MacDonald, A.; Keller, E.A. 1987. Stream channel response to the
removal of large woody debris, Larry Damm Creek, northwestern
California. In: Erosion and sedimentation in the Pacific Rim. Publ. no.
165. International Association of Hydrological Sciences; 405-406.
MacDonald, J.S.; Miller, G.; Stewart, R.A. 1988. The effects of logging,
other forest industries and forest management practices on fish: an
initial bibliography. Publ. no. 1622. Canadian Technical Report of
Fisheries and Aquatic Sciences; 212 p.
MacDonald, L.H.; Smart, A.W.; Wissmar, R.C. 1991. Monitoring guidelines
to evaluate effects of forestry activities on streams in the Pacific
Northwest and Alaska. EPA/910/9-91-001. U.S. Environmental
Protection Agency, Water Division.
Mackay, D.M.; Smith, L.A. 1990. Agricultural chemicals in groundwater:
monitoring and management in California. Journal of Soil and
Water Conservation 45(2): 253-255.
108
MacQuarrie, K.T.B.; Sudicky, E.A.; Frind, E.O. 1990. Simulation of
biodegradable organic contaminants in groundwater, 1, numerical
formulation in principal directions. Water Resources Research 26(2):
207-222.
Madej, M.A. 1982. Sediment routing and channel changes in an
aggrading stream, Puget Lowland, Washington. In: Sediment budgets
and routing in forest catchments. Gen. Tech. Rep. PNW-141. Portland,
OR: Pacific Northwest Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 97-108.
Madej, M.A. 1987. Residence times of channel-stored sediment in
Redwood Creek, northwestern California. In: Erosion and
sedimentation in the Pacific Rim. Publ. no. 165. International Association
of Hydrological Sciences; 429-438.
Mahacek-King, V.L.; Shelton, M.L. 1987. Timber harvesting and the
hydrologic response of Redwood Creek, California. Physical
Geography 8(3): 241-256.
Mahendrappa, M.K.; Foster, N.W.; Weetman, G.F.; Krause, H.H. 1986.
Nutrient cycling and availability in forest soils. Canadian Journal of
Soil Science 66: 547-572.
Mahmood, K. 1987. Reservoir sedimentation: impact, extent, and
mitigation. Washington, DC: The World Bank; 128 p.
Majumdar, S.K.; Miller, E.W.; Brenner, F.J. 1987. Environmental
consequences of energy production: problems and prospects. Easton,
PA: Pennsylvania Academy of Sciences; 531 p.
Mann, L.K.; Johnson, D.W.; West, D.C.; Cole, D.W.; Hornbeck, J.W.;
Martin, C.W.; Riekerk, H.; Smith, C.T; Swank, W.T.; Tritton, L.M.;
Van Lear, D.H. 1988. Effects of whole-tree and stem-only clearcutting
on postharvest hydrologic losses, nutrient capital, and regrowth.
Forest Science 34(2): 412-428.
Manning, R.E. 1985. Crowding norms in backcountry settings: a review
and synthesis. Journal of Leisure Research 17(2): 75-89.
Marcus, M.D.; Young, M.K.; Noel, L.E.; Mullen, B.A. 1990. Salmonid-
habitat relationships in the western United States: a review and
indexed bibliography. Gen. Tech. Rep. RM-188. Fort Collins, CO:
Rocky Mountain Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture.
Marion, J.L.; Merriam, L.C. 1985. Predictability of recreational impact
on soils. Soil Science Society of America Journal 49(3): 751-753.
Mark, D.M.; Church, M. 1977. On the misuse of regression in earth
science. Mathematical Geology 9: 63-75.
Markin, G.P. 1982. Drift of insecticidal spray by cold air drainage
winds in western mountains. Res. Note PSW-360. Berkeley, CA:
Pacific Southwest Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 6 p.
Marlow, C.B.; Pogacnik, T.M. 1986. Cattle feeding and resting patterns
in a foothills riparian zone. Journal of Range Management 39(3):
212-216.
Marron, D.C. 1989. Physical and chemical characteristics of a metal-
contaminated overbank deposit, west-central South Dakota, USA.
Earth Surface Processes and Landforms 14(5): 419-432.
Martin, I.L.; Tinney, E.R. 1962. Logging in west coast watershed shows
no effect on area’s water yield. Timberman 63(5): 46-47.
Maser, C.; Tarrant, R.F.; Trappe, J.M.; Franklin, J.F., eds. 1988. From the
forest to the sea: a story of fallen trees. Gen. Tech. Rep. PNW-229.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture.
Matthews, W.V.G.; Kondolf, G.M. 1987. Changes in sediment storage
over time on a coastal California river. In: Erosion and sedimentation
in the Pacific Rim. Publ. no. 165. International Association of
Hydrological Sciences; 413-414.
Mazurski, K.R. 1986. The destruction of forests in the Polish Sudetes
Mountains by industrial emissions. Forest Ecology and Management
17(4): 303-315.
McBean, E.; Al-Nassri, S. 1987. Migration pattern of de-icing salts
from roads. Journal of Environmental Management 25(3): 231-238.
McCarthy, G.T. 1940. Engineering construction: flood control. Fort
Belvoir, VA: The Engineer School.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
McCarthy, J.W.; Stone, E.L. 1991. Changes in soil water tables following
phosphorus fertilization of young slash pine. Soil Science Society of
America Journal 55(5): 1440-1446.
McCashion, J.D.; Rice, R.M. 1983. Erosion on logging roads in northwestern
California: how much is avoidable? Journal of Forestry 81(1): 23-26.
McClurkin, D.C.; Duffy, P.D.; Ursic, S.J.; Nelson, N.S. 1985. Water
quality effects of clearcutting upper coastal plain loblolly pine
plantations. Journal of Environmental Quality 14(3): 329-332.
McDade, M.H.; Swanson, F.J.; McKee, W.A.; Franklin, J.F.; Van Sickle,
J. 1990. Source distances for coarse woody debris entering small
streams in western Oregon and Washington. Canadian Journal of
Forest Research 20:326-330.
McDonald, C.D.; Hammitt, W.E. 1983. Managing river environments
for the participation motives of stream floaters. Journal of
Environmental Management 16(4): 369-378.
McGee, A.B.; Levy, E. 1988. Herbicide use in forestry: communication
and information gaps. Journal of Environmental Management 26(2):
111-126.
McGinnies, W.J. 1987. Effects of hay and straw mulches on the
establishment of seeded grasses and legumes on rangeland and a
coal strip mine. Journal of Range Management 40(2): 119-121.
McGreal, W.S.; Craig, D. 1977. Mass movement activity: an illustration
of differing responses to groundwater conditions from two sites in
Northern Ireland. Irish Geogr. 10: 28-35.
McLean, D.G.; Tassone, B.L. 1991. A sediment budget of the lower Fraser
River. In: Proceedings of the fifth federal interagency sedimentation
conference; 1991 March 18-21; Las Vegas, NV. Washington, DC: Federal
Energy Regulatory Commission; 2-33 to 2-40.
McLeay, D.J.; Birtwell, I.K.; Hartman, G.F.; Ennis, G.L. 1987. Responses
of Arctic grayling (Thymallus arcticus) to acute and prolonged
exposure to Yukon placer mining sediment. Canadian Journal of
Fisheries and Aquatic Sciences 44(3): 658-673.
McNabb, D.H.; Gaweda, F.; Froehlich, H.A. 1989. Infiltration, water
repellency, and soil moisture content after broadcast burning a
forest site in southwest Oregon. Journal of Soil and Water Conservation
44(1): 87-90.
McNabb, D.H.; Swanson, F.J. 1990. Effects of fire on soil erosion. In:
Natural and prescribed fire in Pacific Northwest forests. Corvallis, OR:
Oregon State University Press; 159-176.
McNeil, W.J.; Ahnell, W.H. 1964. Success of pink salmon spawning
relative to size of spawning bed materials. Special Science Report -
Fish 469. Washington, DC: U.S. Department of Interior, Fish and
Wildlife Service; 15 p.
Meehan, W.R.; Brusven, M.A.; Ward, J.F. 1987. Effects of artificial
shading on distribution and abundance of juvenile chinook salmon
(Oncorhynchus tshawytscha). Great Basin Naturalist 47(1): 22-31.
Meehan, W.R.; Farr, W.A.; Bishop, D.M.; Patric, J.H. 1969. Some effects of
clearcutting on salmon habitat of two southeast Alaska streams. Res.
Paper PNW-82. Portland, OR: Pacific Northwest Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture; 45 p.
Meehan, W.R.; Swanston, D.N. 1977. Effects of gravel morphology on
fine sediment accumulation and survival of incubating salmon
eggs. Res. Paper PNW-220. Portland, OR: Pacific Northwest Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 16 p.
Megahan, W.F. 1972. Subsurface flow interception by a logging road in
mountains of central Idaho. In: National symposium on watersheds
in transition. Fort Collins, CO: American Water Resources Association
and Colorado State University; 350-356.
Megahan, W.F. 1974. Erosion over time on severely disturbed granitic
soils: a model. Res. Paper INT-156. Ogden, UT: Intermountain Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 14 p.
Megahan, W.F. 1975. Sedimentation in relation to logging activities in
the mountains of central Idaho. In: Prospective technology for
predicting sediment yields and sources. ARS-S-40. Washington, DC:
U.S. Department of Agriculture, Agricultural Research Service; 74-82.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Megahan, W.F. 1978. Erosion processes on steep granitic road fills in
central Idaho. Soil Science Society of America Journal 42(2): 350-357.
Megahan, W.F. 1979. Channel sediment storage behind obstructions in
some forested catchments draining granitic bedrock of the Idaho
Batholith. In: Proceedings of the symposium on sediment budgets and
routing in forest catchments. Gen. Tech. Rep. PNW-141. Portland, OR:
Pacific Northwest Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture.
Megahan, W.F. 1980. Erosion from roadcuts in granitic slopes of the Idaho
Batholith. Abstract from Cordilleran Section of the Geological Society of
America, 76th Annual Meeting, Oregon State University, Corvallis.
Megahan, W.F. 1983. Hydrologic effects of clearcutting and wildfire on
steep granitic slopes in Idaho. Water Resources Research 19(3): 811-819.
Megahan, W.F.; Bohn, C.C. 1989. Progressive, long-term slope failure
following road construction and logging on noncohesive granitic
soils of the Idaho Batholith. In: Headwaters hydrology. Bethesda,
MD: American Water Resources Association; 501-510.
Megahan, W.F.; Kidd, W.J., Jr. 1972a. Effects of logging and logging
roads on erosion and sediment deposition from steep terrain. Journal
of Forestry 70: 136-141.
Megahan, W.F.; Kidd, W.J., Jr. 1972b. Effects of logging roads on
sediment production rates in the Idaho Batholith. Res. Paper INT-
123. Ogden, UT: Intermountain Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 14 p.
Megahan, W.F.; Monsen, S.B.; Wilson, M.D. 1991. Probability of sediment
yields from surface erosion on granitic roadfills in Idaho. Journal of
Environmental Quality 20(1): 53-60.
Megahan, W.F.; Nowlin, R.A. 1976. Sediment storage in channels draining
small forested watersheds in the mountains of central Idaho. In:
Proceedings of the third federal inter-agency sedimentation conference;
1976 March 22-25; Denver CO. Washington, DC: Water Resources
Council, Sedimentation Committee; 4-115 to 4-126.
Megahan, W.F.; Seyedbagheri, K.A.; Dodson, P.C. 1983. Long-term erosion
on granitic roadcuts based on exposed tree roots. Earth Surface
Processes 8(1): 19-28.
Megahan, W.F.; Seyedbagheri, K.A; Mosko, T.L.; Ketcheson, G.L. 1986.
Construction phase sediment budget for forest roads on granitic
slopes in Idaho. In: Drainage basin sediment delivery. Publ. no. 159.
International Association of Hydrological Sciences; 31-39 I.
Melton, M.A. 1965. The geomorphic and paleoclimatic significance of
alluvial deposits in southern Arizona. Journal of Geology 73:1-38.
Mersereau, R.C.; Dyrness, C.T. 1972. Accelerated mass wasting after
logging and slash burning in western Oregon. Journal of Soil and
Water Conservation 27(3): 112-114.
Meyer, G.J.; Schoeneberger, P.J.; Huddleston, J.H. 1975. Sediment yields
from roadsides: an application of the Universal Soil Loss Equation.
Journal of Soil and Water Conservation 30(6): 289-291.
Michael, J.H., Jr. 1989. Life history of anadromous cutthroat trout in Snow
and Salmon Creeks, Jefferson County, Washington, with implications
for management. California Fish and Game 75(4): 188-203.
Michael, J.L.; Neary, D.G.; Wells, M.J.M. 1989. Picloram movement in
soil solution and streamflow from a Coastal Plain forest. Journal of
Environmental Quality 18: 89-95.
Miles, J.R.W. 1976. Insecticide residues on stream sediments in Ontario,
Canada. Pesticide Monitoring Journal 10(3): 87-91.
Miller, A.J. 1990. Fluvial response to debris associated with mass
wasting during extreme floods. Geology 18(7): 599-602.
Miller, J.H.; Bace, A.C. 1980. Streamwater contamination after aerial
application of pelletized herbicide. Res. Note SO-255. New Orleans,
LA: Southern Forest Experiment Station, Forest Service, U.S.
Department of Agriculture; 4 p.
Miller, J.H.; Sirois, D.L. 1986. Soil disturbance by skyline yarding vs.
skidding in a loamy hill forest. Soil Science Society of America
Journal 50: 1579.
Miller, K.R.; Reid, W.V.; Barber, C.V. 1991. Deforestation and species
loss. In: Mathews, J.T., ed. Preserving the global environment. New
York: W.W. Norton and Co.; 78-111.
109
Milne, M.M.; Young, D.W. 1989. The impact of stockwatering ponds
(stockponds) on runoff from large Arizona watersheds. Water
Resources Bulletin 25(1): 165-174.
Mitchell, J.E.; Rodgers, R.T. 1985. Food habits and distribution of
cattle on a forest and pasture range in northern Idaho. Journal of
Range Management 38(3): 214-219.
Mitchell, M.J.; Driscoll, C.T.; Fuller, R.D; David, M.B.; Likens, G.E.
1989. Effect of whole-tree harvesting on the sulfur dynamics of a
forest soil. Soil Science Society of America Journal 53(3): 933-940.
Mitchell, R.G.; Waring, R.H.; Pitman, G.B. 1983. Thinning lodgepole
pine increases tree vigor and resistance to mountain pine beetle.
Forest Science 29:204-211.
Moehring, D.M.; Rawls, I.W. 1970. Detrimental effects of wet weather
logging. Journal of Forestry 68: 166-167.
Moglen, G.E.; McCuen, R.H. 1988. Effects of detention basins on in-
stream sediment movement. Journal of Hydrology 104(1-4): 129-
140.
Mok, W.M.; Riley, J.A.; Wai, C.M. 1988. Arsenic speciation and quality
of groundwater in a lead-zinc mine. Water Research 22: 769-774.
Mok, W.M.; Wai, C.M. 1989. Distribution and mobilization of arsenic
species in the creeks around the Blackbird mining district, Idaho.
Water Research 23(1): 7-14.
Monti, P.; Mackintosh, E.E. 1979. Effect of camping on surface soil
properties in the boreal forest region of northwestern Ontario,
Canada. Soil Science Society of America Journal 43: 1024-1029.
Moody, D.W. 1990. Groundwater contamination in the United States.
Journal of Soil and Water Conservation 45(2): 170-179.
Moore, I.D.; Burch, G.J. 1986. Sediment transport capacity of sheet and
rill flow: application of unit stream power theory. Water Resources
Research 22(8): 1350-1360.
Moore, K.M.S.; Gregory, S.V. 1988a. Response of young-of-the-year cutthroat
trout to manipulation of habitat structure in a small stream. Transactions
of the American Fisheries Society 117(2): 162-170.
Moore, K.M.S.; Gregory, S.V. 1988b. Summer habitat utilization and
ecology of cutthroat trout fry (Salmo clarki) in Cascade Mountain
streams. Canadian Journal of Fisheries and Aquatic Sciences 45(11):
1921-1930.
Moore, P.D. 1975. Origin of blanket mires. Nature 256:267-269.
Morantz, D.L.; Sweeney, R.K.; Shirvell, C.S.; Longard, D.A. 1987. Selection
of microhabitat in summer by juvenile Atlantic salmon (Salmo salar).
Canadian Journal of Fisheries and Aquatic Sciences 44(1): 120-129.
Morris, L.A.; Pritchett, W.L.; Swindel, B.F. 1983. Displacement of nutrients
into windrows during site preparation of a flatwood forest. Soil
Science Society of America Journal 47:591-594.
Mosley, M.P. 1980. The impact of forest road erosion in the Dart
Valley, Nelson, New Zealand. Journal of Forestry 25(2): 184-198.
Mosley, M.P. 1981. The influence of organic debris on channel
morphology and bedload transport in a New Zealand forest stream.
Earth Surface Processes and Landforms 6(6): 571-580.
Motto, H.L.; Daines, R.H.; Chilko, D.M.; Motto, C.K. 1970. Lead in soils
and plants. Its relation to traffic volume and proximity to highways.
Environmental Science and Technology 4:231-238.
Munro, C.H.; Foster, D.N.; Nelson, R.C.; Bell, F.C. 1967. The Georges
River hydraulic, hydrologic and reclamation studies. Publ. no. 101.
University of New South Wales Water Research Laboratory.
Murgatroyd, A.L.; Ternan, J.L. 1983. The impact of afforestation on
stream bank erosion and channel form. Earth Surface Processes and
Landforms 8(4): 357-370.
Murphy, A.H. 1970. Predicted forage yield based on fall precipitation
in California annual grasslands. Journal of Range Management 23:
363-365.
Murphy, M.L.; Hall, J.D. 1981. Varied effects of clear-cut logging on
predators and their habitat in small streams of the Cascade
Mountains, Oregon. Canadian Journal of Fisheries and Aquatic Sciences
38:137-145.
110
Murphy, M.L.; Heifetz, J.; Johnson, S.W.; Koski, K.V.; Thedinga, J.F.
1986. Effects of clear-cut logging with and without buffer strips on
juvenile salmonids in Alaskan streams. Canadian Journal of Fisheries
and Aquatic Sciences 43(8): 1521-1533.
Nadler, C.T.; Schumm, S.A. 1981. Metamorphosis of South Platte and
Arkansas Rivers, eastern Colorado. Physical Geography 2(2): 95-115.
Nanson, G.C.; Young, R.W. 1981. Downstream reduction of rural channel
size with contrasting urban effects in small coastal streams of
southeastern Australia. Journal of Hydrology 52:239-255.
Naveh, Z.; Lieberman, A.S. 1984. Landscape ecology, theory and
application. New York: Springer-Verlag; 356 p.
NCASI. 1984. Forest management practices and cumulative effects on
water quality and utility. Tech. Bull. 435. National Council of the
Paper Industry for Air and Stream Improvement (NCASI); 48 p.
NCASI. 1985. Catalog of landslide inventories for the Northwest. Tech.
Bull. 456. National Council of the Paper Industry for Air and Stream
Improvement (NCASI); 78 p.
NCASI. 1986. Papers Presented at the American Geophysical Union
Meeting on Cumulative Effects. Tech. Bull. 490. National Council of
the Paper Industry for Air and Stream Improvement (NCASI); 73 p.
Nearing, M.A.; Foster, G.R.; Lane, L.J.; Finkner, S.C. 1989. A process-
based soil erosion model for U.S.D.A. - Water erosion prediction
project technology. Transactions of the American Society of Agricultural
Engineers 32(5): 1587-1593.
Neary, D.G.; Bush, P.B.; Douglass, J.E.; Todd, R.L. 1985. Picloram
movement in an Appalachian hardwood forest watershed. Journal
of Environmental Quality 14: 585-592.
Neary, D.G.; Michael, J.L. 1989. Effect of sulfometuron methyl on
ground water and stream quality in coastal plain forest watersheds.
Water Resources Bulletin 25(3): 617-624.
Neller, R.J. 1988. A comparison of channel erosion in small urban and
rural catchments, Armidale, New South Wales. Earth Surface
Processes and Landforms 13(1): 1-8.
Neuman, D.R.; Munshower, F.F. 1984. Copper and molybdenum uptake
by forages grown on coal mine soils. Journal of Range Management
37(6): 517-520.
Ng, H.Y.F.; Marsalek, J. 1989. Simulation of the effects of urbanization
on basin streamflow. Water Resources Bulletin 25(1): 117-124.
Nichols, D.G.; Scholz, A.T. 1989. Concentrations of Cd, Sr and U in
fish and water samples collected from a small stream receiving
uranium mine drainage. Journal of Freshwater Ecology 5(1): 13-26.
Nielsen, J. Pacific Southwest Forest and Range Experiment Station. Forest
Service, U.S. Department of Agriculture, Berkeley, CA: [Personal
communication with Leslie Reid]. 1991.
Nihlgard, B. 1985. The ammonium hypothesis: an additional explanation
to the forest dieback in Europe. Ambio 14(1): 2-9.
Noggle, C.C. 1978. Behavioral, physiological and lethal effects of
suspended sediment on juvenile salmonids. Seattle, WA: University
of Washington; 87 p. MS Thesis.
Nolan, K.M.; Janda, R.J. 1981. Use of short-term water and suspended-
sediment discharge observations to assess impacts of logging on
stream-sediment discharge in the Redwood Creek basin,
northwestern California, USA. In: Erosion and sediment transport in
Pacific Rim steeplands. Publ. no. 132. International Association of
Hydrological Sciences; 415-437.
Norris, L.A.; Lorz, H.W.; Gregory, S.V. 1983. Influence of forest and
rangeland management on anadromous fish habitat in western
North America - 9. Chemicals. Gen. Tech. Rep. PNW-149. Portland,
OR: Pacific Northwest Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 95 p.
Norris, L.A.; Montgomery, M.L.; Johnson, E.R. 1977. The persistence of
2,4,5-T in a Pacific Northwest forest. Weed Science 25:417-422.
Northrup, W.L. 1965. Republican River channel deterioration. In:
Proceedings of the federal inter-agency sedimentation conference; 1963
Jan. 28-Feb. 1; Jackson, MS. Misc. Pub. 970. Washington, DC: U.S.
Department of Agriculture; 409-424.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Nouh, M. 1988. Methods of estimating bed load transport rates applied
to ephemeral streams. In: Sediment budgets. Publ. no. 174. International
Association of Hydrological Sciences; 107-115.
Novotny, V.; Chesters, G. 1989. Delivery of sediment and pollutants
from nonpoint sources: a water quality perspective. Journal of Soil
and Water Conservation 44(6): 568-576.
Nutter, W.L.; Tkacs, T.; Bush, P.B.; Neary, D.G. 1984. Simulation of
herbicide concentrations in stormflow from forested watersheds.
Water Resources Bulletin 20(6): 851-858.
Oliver, C.D. 1981. Forest development in North America following
major disturbances. Forest Ecology and Management 3(3): 153-168.
O’Loughlin, C.L. 1972. A preliminary study of landslides in the Coast
Mountains of Southwest British Columbia. In: Slaymaker, H.O.;
McPherson, H.J., eds. Mountain geomorphology. Vancouver, BC:
Tantalus Research Ltd.; 101-111.
Olson, K.C.; White, R.S.; Sindelar, B.W. 1985. Response of vegetation of
the northern Great Plains to precipitation amount and grazing
intensity. Journal of Range Management 38(4): 357-361.
Omart, C.P.; Williams, C.B., Jr. 1979. The effects of photochemical oxidants
on radial growth increment for five species of conifers in the San
Bernardino National Forest. Plant Disease Reporter 63:1038-1042.
Orme, A.R.; Bailey, R.G. 1971. Vegetation conversion and channel
geometry in Monroe Canyon, California. Yearbook of Pacific Coast
Geographers 33:65-82.
Osborne, L.L.; Wiley, M.J. 1988. Empirical relationships between land
use/cover and stream water quality in an agricultural watershed.
Journal of Environmental Management 26: 9-27.
Ottaway, E.M.; Clarke, A. 1981. A preliminary investigation into the
vulnerability of young trout (Salmo trutta L.) to downstream
displacement by high water velocity. Journal of Fish Biology 19:
135-145.
Owe, M. 1985. Long-term streamflow observations in relation to basin
development. Journal of Hydrology 78: 243-260.
Packer, P.E. 1953. Effects of trampling disturbance on watershed
condition, runoff and erosion. Journal of Forestry 51: 28-31.
Padilla, F.; Lafrance, P.; Robert, C.; Villeneuve, J.P. 1988. Modeling the
transport and the fate of pesticides in the unsaturated zone considering
temperature effects. Ecological Modelling 44(1-2): 73-88.
Page-Dumroese, D.S.; Jurgensen, M.F.; Graham, R.T.; Harvey, A.E. 1987.
Soil physical and chemical properties associated with three site
preparation techniques and their effects on growth and survival of
seedlings in northern Idaho. Gen. Tech. Rep. SE-46. Asheville, NC:
Southeastern Forest Experiment Station, Forest Service, U.S. Department
of Agriculture; 66-70.
Parker, A.J. 1986. Persistence of lodgepole pine forests in the central
Sierra Nevada. Ecology 67(6): 1560-1567.
Parker, R.S. 1988. Uncertainties in defining the suspended sediment
budget for large drainage basins. In: Sediment budgets. Publ. no.
174. International Association of Hydrological Sciences; 523-532.
Parsons, A.J. 1982. Slope profile variability in first-order drainage
basins. Earth Surface Processes and Landforms 7(1): 71-78.
Parsons, D.J.; DeBenedetti, S.H. 1979. Impact of fire suppression on a
mixed conifer forest. Forest Ecology and Management 2: 21-33.
Pasche, E.; Rouve, G. 1985. Overbank flow with vegetatively roughened
flood plains. Journal of American Society of Civil Engineers, Hydraulics
Division 111(9): 1262-1278.
Patric, J.H. 1976. Soil erosion in the eastern forest. Journal of Forestry
74(10): 671-677.
Patric, J.H.; Gorman, J.L. 1978. Soil disturbance caused by skyline cable
logging on steep slopes in West Virginia. Journal of Soil and Water
Conservation 33(1): 32-35.
Patric, J.H.; Reinhart, K.G. 1971. Hydrologic effects of deforesting two
mountain watersheds in West Virginia. Water Resources Research
7: 1182-1188.
Payne, G.F.; Foster, J.W.; Leininger, W.C. 1983. Vehicle impacts on
northern Great Plains range vegetation. Journal of Range Management
36(3): 327-331.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Pearce, A.J. 1976. Geomorphic and hydrologic consequences of
vegetation, Sudbury, Ontario. Canadian Journal of Earth Sciences
13: 1358-1373.
Pearson, R.G.; Jones, N.V. 1975. The effects of dredging operations on
the benthic community of a chalk stream. Biological Conservation
8(4): 273-278.
Pemberton, E.L. 1976. Channel changes in the Colorado River below
Glen Canyon Dam. In: Proceedings of the third federal inter-agency
sedimentation conference; 1976 March 22-25; Denver, CO. Washington,
DC: Water Resources Council, Sedimentation Committee; 5-61 to 5-73.
Penman, H.L. 1948. Natural evaporation from open water, soil, and
grass. In: Proceedings of the Royal Society of London, series A 193:
120-145.
Pennell, K.D.; Hornsby, A.G.; Jessup, R.E.; Rao, P.S.C. 1990. Evaluation
of five simulation models for predicting aldicarb and bromide
behavior under field conditions. Water Resources Research 26(11):
2679-2694.
Perfect, E.; Kay, B.D.; van Loon, W.K.P.; Sheard, R.W.; Pojasok, T. 1990.
Rates of change in soil structural stability under forages and corn.
Soil Science Society of America Journal 54(1): 179-185.
Perkins, S.J. 1989. Landslide deposits in low-order streams— their
erosion rates and effects on channel morphology. In: Headwaters
hydrology. Bethesda, MD: American Water Resources Association;
173-182.
Peterson, D.L.; Arbaugh, M.J.; Wakefield, V.A.; Miller, P.R. 1987a.
Evidence of growth reduction in ozone-injured Jeffrey Pine (Pinus
jeffreyi Grev. and Balf.) in Sequoia and Kings Canyon National
Park. JAPCA 37(8): 906-912.
Peterson, E.B.; Chan, Y.-H.; Peterson, N.M; Constable, G.A.; Caton, R.B.;
Davis, C.S.; Wallace, R.R.; Yarranton, G.A. 1987b. Cumulative effects
assessment in Canada: an agenda for action and research. Canadian
Environmental Assessment Research Council; 66 p.
Peterson, N.P. 1982a. Population characteristics of juvenile coho salmon
(Oncorhynchus kisutch) overwintering in riverine ponds. Canadian
Journal of Fisheries and Aquatic Sciences 39(9): 1303-1307.
Peterson, N.P. 1982b. Immigration of juvenile coho salmon
(Oncorhynchus kisutch) into riverine ponds. Canadian Journal of
Fisheries and Aquatic Sciences 39(9): 1308-1310.
Peterson, N.P.; Reid, L.M. 1984. Wall-base channels: their evolution,
distribution, and use by juvenile coho salmon in the Clearwater
River, Washington. In: Walton, J.M.; Houston, J.B., eds. Proceedings
of the Olympic wild fish conference; 1983 March 23-25; Port Angeles,
WA; 215-225.
Petryk, S.; Bosmajian, G., III. 1975. Analysis of flow through vegetation.
Journal of American Society of Civil Engineers, Hydraulics Division
101(7): 871-884.
Petts, G.E. 1977. Channel response to flow regulation: the case of the
River Derwent, Derbyshire. In: Gregory, K.J., ed. River channel
changes. New York: John Wiley and Sons; 145-164.
Petts, G.E. 1979. Complex response of river channel morphology to
reservoir construction. Progress in Physical Geography 3: 329-362.
Petts, G.E. 1984. Sedimentation within a regulated river. Earth Surface
Processes and Landforms 9(2): 125-134.
Pfister, R.D. 1969. Effect of roads on growth of western white pine
plantations in northern Idaho. Res. Paper INT-65. Ogden, UT:
Intermountain Forest and Range Experiment Station, Forest Service,
U.S. Department of Agriculture; 8 p.
Pfister, R.E. 1977. Campsite choice behavior in the river setting: a pilot
study on the Rogue River, Oregon. In: Proceedings of the river
recreation management and research symposium. Gen. Tech. Rep. NC-
28. St. Paul, MN: North Central Forest Experiment Station, Forest
Service, U.S. Department of Agriculture; 351-358.
Phillips, J.D. 1990. Fluvial sediment budgets in the North Carolina
Piedmont. Geomorphology 4(3/4): 231-242.
Phillips, R.W. 1970. Effects of sediment on the gravel environment
and fish production. In: Proceedings of a symposium on forest land
uses and the stream environment. Corvallis, OR: Oregon State
University; 64-74.
111
Pickup, G. 1980. Hydrologic and sediment modelling studies in the
environmental impact assessment of a major tropical dam project.
Earth Surface Processes 5(1): 61-75.
Piehl, B.T.; Beschta, R.L.; Pyles, M.R. 1988a. Ditch-relief culverts and
low-volume forest roads in the Oregon Coast Range. Northwest
Science 62(3): 91-98.
Piehl, B.T.; Pyles, M.R.; Beschta, R.L. 1988b. Flow capacity of culverts
on Oregon Coast Range forest roads. Water Resources Bulletin
24(3): 631-637.
Pierce, R.S.; Lull, H.W.; Storey, H.C. 1958. Influence of land use and
forest condition on soil freezing and snow depth. Forest Science 4:
246-263.
Pitelka, L.F.; Raynal, D.J. 1989. Forest decline and acidic deposition.
Ecology 70(1): 2-10.
Pizzuto, J.E.; Meckelnburg, T.S. 1989. Evaluation of a linear bank
erosion equation. Water Resources Research 25(5): 1005-1013.
Platou, K.A.; Tueller, P.T. 1985. Evolutionary implications for grazing
management systems. Rangelands 7(2): 57-61.
Platts, W.S. 1981. Influence of forest and rangeland management on
anadromous fish habitat in western North America. 7. Effects of
livestock grazing. Gen. Tech. Rep. PNW-124. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture.
Platts, W.S.; Martin, S.B.; Primbs, E.R.J. 1979. Water quality in an
Idaho stream degraded by acid mine waters. Gen. Tech. Rep. INT-
67. Ogden, UT: Intermountain Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 19 p.
Platts, W.S.; Nelson, R.L. 1988. Fluctuations in trout populations and
their implications for land-use evaluation. North American Journal
of Fisheries Management 8(3): 333-345.
Platts, W.S.; Nelson, R.L.; Casey, O.; Crispin, V. 1983. Riparian-stream
habitat conditions on Tabor Creek, Nevada, under grazed and
ungrazed conditions. In: Western association of fish and wildlife
agencies annual conference proceedings, 63rd; Teton Village, WY:
The Association; 162-174.
Platts, W.S.; Torquemada, R.J.; McHenry, M.L.; Graham, C.K. 1989.
Changes in salmon spawning and rearing habitat from increased
delivery of fine sediment to the South Fork Salmon River, Idaho.
Transactions of the American Fisheries Society 118: 274-283.
Pope, C.A., Jr.; Jones, J.W. 1990. Value of wilderness designation in
Utah. Journal of Environmental Management 30(2): 157-174.
Potts, D.F. 1984. Hydrologic impacts of a large-scale mountain pine
beetle (Dendroctonus ponderosae Hopkins) epidemic. Water Resources
Bulletin 20(3): 373-378.
Potts, D.F.; Peterson, D.L.; Zuuring, H.R. 1989. Estimating postfire
water production in the Pacific Northwest. Res. Paper PSW-197.
Berkeley, CA: Pacific Southwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 9 p.
Power, J.F.; Sandoval, F.M.; Ries, R.E.; Merrill, S.D. 1981. Effects of
topsoil and subsoil thickness on soil water content and crop
production on a disturbed soil. Soil Science Society of America
Journal 45(1): 124-128.
Power, M.E. 1990. Resource enhancement by indirect effects of grazers:
armored catfish, algae, and sediment. Ecology 71(3): 897-904.
Prestegaard, K.L. 1988. Morphological controls on sediment delivery
pathways. In: Sediment budgets. Publ. no. 174. International Association
of Hydrological Sciences; 533-540.
Price, A.G.; Watters, R.J. 1989. The influence of the overstory, understory
and upper soil horizons on the fluxes of some ions in a mixed
deciduous forest. Journal of Hydrology 109(1/2): 185-198.
Prigogine, I.; Stengers, I. 1984. Order out of chaos. New York: Bantam
Books; 349 p.
Radford, D.S.; Hartland-Rowe, R. 1972. Some conservational problems
posed by hydroelectric power installations in Alberta. Biological
Conservation 4(3): 166-168.
Ragheb, M.G. 1980. Interrelationships among leisure participation,
leisure satisfaction and leisure activities. Journal of Leisure Research
12(2): 138-149.
112
Rango, A. 1970. Possible effects of precipitation modification on stream
channel geometry and sediment yield. Water Resources Research 6:
1765-1770.
Rathjens, G.W. 1991. Energy and climatic change. In: Mathews, J.T., ed.
Preserving the global environment. New York: W.W. Norton and Co.;
154-186.
Redding, J.M.; Schreck, C.B.; Everest, F.H. 1987. Physiological effects
on coho salmon and steelhead of exposure to suspended solids.
Transactions of the American Fisheries Society 116(5): 737-744.
Reeves, F.B.; Wagner, D.; Moorman, T.; Kiel, J. 1979. The role of
endomycorrhizae in revegetation practices in the semi-arid west. I.
A comparison of incidence of mycorrhizae in severely disturbed vs.
natural environments. American Journal of Botany 66: 6-13.
Reeves, G.H.; Everest, F.H.; Hall, J.D. 1987. Interactions between the
redside shiner (Richardsonius baltectus) and the steelhead trout
(Salmo gairdneri) in western Oregon: the influence of water
temperature. Canadian Journal of Fisheries and Aquatic Sciences
44(9): 1603-1613.
Reeves, G.H.; Everest, F.H.; Nickelson, T.E. 1989. Identification of physical
habitats limiting the production of coho salmon in western Oregon
and Washington. Gen. Tech. Rep. PNW-245. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 18 p.
Reid, L.M. 1989. Channel incision by surface runoff in grassland
catchments. Seattle, WA: University of Washington; 199 p. Dissertation.
Reid, L.M. 1990. Construction of sediment budgets to assess erosion in
Shinyanga Region, Tanzania. In: Research needs and applications to
reduce erosion and sedimentation in tropical steeplands. Publ. no. 192.
International Association of Hydrological Sciences; 105-114.
Reid, L.M.; Dunne, T. 1984. Sediment production from forest road
surfaces. Water Resources Research 20(11): 1753-1761.
Reid, L.M.; Dunne, T.; Cederholm, C.J. 1981. Application of sediment
budget studies to the evaluation of logging road impact. Journal of
Hydrology (N.Z.) 20(1): 49-62.
Reid, W.V.; Miller, K.R. 1989. Keeping options alive: the scientific
basis for conserving biodiversity. Washington, DC: World Resources
Institute; 128 p.
Reid, W.V.; Trexler, M.C. 1991. Drowning the national heritage: climatic
change and U.S. coastal biodiversity. Washington, DC: World
Resources Institute; 48 p.
Reinhart, K.G. 1964. Effect of a commercial clearcutting in West Virginia
on overland flow and storm runoff. Journal of Forestry 62:167-171.
Reisner, M. 1986. Cadillac desert. New York: Penguin Books; 582 p.
Renard, K.G.; Foster, G.R. 1985. Managing rangeland soil resources:
the universal soil loss equation. Rangelands 7(3): 118-122.
Reynolds, B. 1986. A comparison of element outputs in solution,
suspended sediments and bedload for a small upland catchment.
Earth Surface Processes and Landforms 11(2): 217-222.
Reynolds, J.B.; Simmons, R.C.; Burkholder, A.R. 1989. Effects of placer
mining discharge on health and food of Arctic grayling. Water
Resources Bulletin 25(3): 625-636.
Rice, R.M. 1985. A method of identifying landslide hazard. In: Abstracts
to presentations at the 1985 West Coast regional meeting. National
Council of the Paper Industry for Air and Stream Improvement; C-1
to C-5.
Rice, R.M.; Datzman, P.A. 1981. Erosion associated with cable and
tractor logging in northwestern California. In: Erosion and sediment
transport in Pacific Rim steeplands. Publ. no. 132. International
Association of Hydrological Sciences; 362-374.
Rice, R.M.; Gradek, P.D. 1984. Limits on the usefulness of erosion-
hazard ratings: experiences in northwestern California. Canadian
Journal of Forest Resources 14:559-564.
Rice, R.M.; Lewis, J. 1986. Identifying unstable sites on logging roads.
In: Proceedings of Division 1. 18th International Union of Forestry
Research Organizations World Congress, vol. 1. IUFRO; 239-247.
Rice, R.M.; McCashion, J.D. 1985. Site conditions related to erosion on
logging roads. In: Proceedings of the international symposium on
erosion, debris flow and disaster prevention; 1985 Sept. 3-5; Tsukuba,
Japan; 69-74.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Rice, R.M.; Pillsbury, N.H. 1982. Predicting landslides in clearcut patches.
In: Recent developments in the explanation and prediction of erosion
and sediment yield. Publ. no. 137. International Association of
Hydrological Sciences; 303-311.
Rice, R.M.; Pillsbury, N.H.; Schmidt, K.W. 1985. A risk analysis approach
for using discriminant functions to manage logging-related landslides
on granitic terrain. Forest Science 31(3): 772-784.
Rice, R.M.; Tilley, F.B.; Datzman., P.A. 1979. A watershed’s response to
logging and roads: South Fork of Caspar Creek, California, 1967-
1976. Res. Paper PSW-146. Berkeley, CA: Pacific Southwest Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 12 p.
Rice, R.M.; Wallis, J.R. 1962. How a logging operation can affect
streamflow. Forest Industries 89(11): 38-40.
Richards, K.; Greenhalgh, C. 1984. River channel change: problems of
interpretation illustrated by the river Derwent, North Yorkshire.
Earth Surface Processes and Landforms 9(2): 175-180.
Richey, J.E.; Perkins, M.A.; Goldman, C.R. 1975. Effects of kokanee
salmon (Oncorhynchus nerka) decomposition on the ecology of a
subalpine stream. Journal of the Fisheries Research Board of Canada
32: 817-820.
Richmond, R.M. 1983. Problems and opportunities of forestland grazing
in the Pacific Northwest. In: Roche, B.F., ed. Forestland grazing.
Proceedings of a symposium held Feb 23-25, 1983; Spokane, WA;
71-73.
Rickard, W.H. 1985. Experimental cattle grazing in a relatively undisturbed
shrubsteppe community. Northwest Science 59(1): 66-72.
Rickard, W.H.; Cushing, C.E. 1982. Recovery of streamside woody
vegetation after exclusion of livestock grazing. Journal of Range
Management 35(3): 360-361.
Riekerk, H. 1985. Water quality effects of pine flatwoods silviculture.
Journal of Soil and Water Conservation 80(3): 306-309.
Riekerk, H. 1989. Influence of silvicultural practices on the hydrology
of pine flatwoods in Florida. Water Resources Research 25(4): 713-
720.
Rippel, P.; Pieper, R.D.; Lymbery, G.A. 1983. Vegetational evaluation of
pinyon-juniper cabling in south-central New Mexico. Journal of
Range Management 36(1): 13-15.
Roath, L.R.; Krueger, W.C. 1982. Cattle grazing influence on a mountain
riparian zone. Journal of Range Management 35(1): 100-103.
Roberts, J.R.; Cherry, J.A.; Schwartz, F.W. 1982. A case study of a
chemical spill: polychlorinated biphenyls (PCBs), 1, History,
distribution, and surface translocation. Water Resources Research
18(3): 525-534.
Roberts, R.G.; Church, M. 1986. The sediment budget in severely
disturbed watersheds, Queen Charlotte Ranges, British Columbia.
Canadian Journal of Forest Research 16:1092-1106.
Roberts, T.M.; Skeffington, R.A.; Blank, L.W. 1989. Causes of type 1
spruce decline in Europe. Forestry 62(3): 179-222.
Roe, A.L.; Amman, G.D. 1970. The mountain pine beetle in lodgepole
pine forests. Res. Paper INT-171. Ogden, UT: Intermountain Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 23 p.
Roehl, J.W. 1962. Sediment source area, delivery ratio, and influencing
morphological factors. Publ. no. 59. International Association of
Scientific Hydrology Committee on Land Erosion; 202-213.
Roels, J.M. 1985. Estimation of soil loss at a regional scale based on
plot measurements—some critical considerations. Earth Surface
Processes and Landforms 10(6): 587-598.
Rollerson, T.P.; Howes, D.E.; Sondheim, M.W. 1986. An approach to
predicting post-logging slope stability for coastal British Columbia.
In: Assessing forest practice rules and managing to avoid landslides
and cumulative effects. Publ. no. 496. National Council of the Paper
Industry on Air and Stream Improvement (NCASI); 27-39.
Rosenau, M.L.; McPhail, J.D. 1987. Inherited differences in agonistic
behavior between two populations of coho salmon. Transactions of
the American Fisheries Society 116(4): 646-654.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Rossi, R.S. 1980. History of cultural influences on the distribution and
reproduction of oaks in California. Proceedings of a symposium on
the ecology, management, and utilization of California Oaks. Gen.
Tech. Rep. PSW-44. Berkeley, CA: Pacific Southwest Forest and
Range Experiment Station, Forest Service, U.S. Department Agiculture;
7-18.
Rothacher, J. 1963. Net precipitation under a Douglas-fir forest. Forest
Science 9: 423-429.
Rothacher, J. 1970. Increases in water yield following clearcut logging
in the Pacific Northwest. Water Resources Research 6(2): 653-658.
Rothacher, J. 1973. Does harvest in west slope Douglas-fir increase
peak flow in small forest streams? Res. Paper PNW-163. Portland,
OR: Pacific Northwest Forest and Range Experiment Station, Forest
Service, U.S. Department of Agriculture; 13 p.
Routledge, R.D. 1987. The impact of soil degradation on the expected
present net worth of future timber harvests. Forest Science 33(4):
823-834.
Russell, G.W.; St. Pierre, R.A. 1987. Impacts of hydropower development
on anadromous fish in the northeast United States. In: Majumdar,
S.K.; Miller, E.W.; Brenner, F.J., eds. Environmental consequences of
energy production: problems and prospects. Easton, PA: Pennsylvania
Academy of Sciences; 319-333.
Ruth, R.H. 1967. Silvicultural effects of skyline crane and high-lead
yarding. Journal of Forestry 65:251-255.
Ruyle, G.B.; Bowns, J.E. 1985. Forage use by cattle and sheep grazing
separately and together on summer range in southwestern Utah.
Journal of Range Management 38(4): 299-302.
Ryan, P.F.; Hornberger, G.M.; Crosby, B.J.; Galloway, J.N.; Webb, J.R.;
Rastetter, E.B. 1989. Changes in the chemical composition of
streamwater in two catchments in the Shenandoah National Park,
Virginia, in response to atmospheric deposition of sulfur. Water
Resources Research 25(10): 2091-2100.
Ryan, S.E.; Grant, G.E. 1991. Downstream effects of timber harvesting
on channel morphology in Elk River Basin, Oregon. Journal of
Environmental Quality 20(f1): 60-72.
Salo, E.O.; Cederholm, C.J. 1981. Cumulative effects of forest
management on watersheds—some aquatic considerations. In:
Standiford, R.B.; Ramacher, S.I., eds. Cumulative effects of forest
management on California watersheds. Publ. no. 3268. University of
California Agricultural Science; 67-78.
Sandecki, M. 1989. Aggregate mining in river systems. California Geology
April 1989: 88-94.
Sanders, L.D.; Walsh, R.G.; Loomis, J.B. 1990. Toward empirical
estimation of the total value of protecting rivers. Water Resources
Research 26(7): 1345-1358.
Sanderson, H.R.; Megnanck, R.A.; Gibbs, K.C. 1986. Range management
and scenic beauty as perceived by dispersed recreationists. Journal
of Range Management 39(5): 464-469.
Sartz, R.S. 1953. Soil erosion on a fire-denuded forest area in the
Douglas fir region. Journal of Soil and Water Conservation 8:279-281.
Saunders, I.; Young, A. 1983. Rates of surface processes on slopes, slope
retreat and denudation. Earth Surface Processes and Landforms 8:
473-501.
Saunders, J.W. 1969. Mass mortalities and behavior of brook trout and
juvenile Atlantic salmon in a stream polluted by agricultural
pesticides. Journal of the Fisheries Research Board of Canada 26(3):
695-699.
Savabi, M.R.; Gifford, G.F. 1987. Application of selected soil loss equations
to trampled soil conditions. Water Resources Bulletin 23(4): 709-716.
Savage, S.M. 1974. Mechanism of fire-induced water repellency in soil.
Soil Science Society of America Proceedings 38(4): 652-657.
Schaetzl, R.J.; Johnson, D.J.; Burns, S.F.; Small, T.W. 1989. Tree uprooting:
review of terminology, process, and environmental implications.
Canadian Journal of Forest Research 19(1): 1-11.
Schertz, D.L.; Moldenhauer, W.C.; Livingston, S.J.; Weesies, G.A.; Hintz,
E.A. 1989. Effect of past soil erosion on crop productivity in Indiana.
Journal of Soil and Water Conservation 44(6): 604-608.
113
Schmidt, R.A.; Troendle, C.A. 1989. Snowfall into a forest and clearing.
Journal of Hydrology 110(3/4): 335-348.
Schreyer, R.; Lime, D.; Williams, D. 1984. Characterizing the influence
of past experience on recreation behavior. Journal of Leisure Research
16:34-50.
Schroeder, W.L.; Brown, G.W. 1984. Debris torrents, precipitation and
roads in two coastal Oregon watersheds. In: O’Loughlin, C.L.; Pearce,
A.J., eds. Symposium on effects of forest land use on erosion and slope
stability. Honolulu, HI: East-West Center, University of Hawaii; 117-
122.
Schulz, T.T.; Leininger, W.C. 1990. Differences in riparian vegetation
structure between grazed areas and exclosures. Journal of Range
Management 43(4): 295-299.
Scott, V.H. 1956. Relative infiltration rates of burned and unburned
upland soils. Transactions. American Geophysical Union 37: 67-69.
Scott, V.H.; Burgy, R.H. 1956. Effects of heat and brush burning on the
physical properties of certain upland soils that influence infiltration.
Soil Science 82: 63-70.
Scott, W.S. 1981. An analysis of factors influencing de-icing salt levels
in streams. Journal of Environmental Management 13:269-287.
Scrimgeour, G.J.; Winterbourn, M.J. 1989. Effects of floods on epilithon
and benthic macroinvertebrate populations in an unstable New
Zealand river. Hydrobiologia 171(1): 33-44.
Scrivener, J.C.; Brownlee, M.J. 1989. Effects of forest harvesting on
spawning gravel and incubation survival of chum (Oncorhynchus
keta) and coho salmon (O. kisutch) in Carnation Creek, British
Columbia. Canadian Journal of Fisheries and Aquatic Sciences 46(4):
681-696.
Searle, M.S.; Jackson, E.L. 1985. Recreation and non-participation and
barriers to participation: considerations for the management of
recreation delivery systems. Journal of Recreation and Park
Administration 3:23-35.
Sedell, J.R.; Froggatt, J.L. 1984. Importance of streamside forests to
large rivers: the isolation of the Willamette River, Oregon, U.S.A.,
from its floodplain by snagging and streamside forest removal.
Verh. Internat. Verein. Limnol. 22: 1828-1834.
Sedell, J.R.; Reeves, G.H.; Hauer, F.R.; Stanford, J.A.; Hawkins, C.P.
1990. Role of refugia in recovery from disturbances: modern
fragmented and disconnected river systems. Environmental
Management 14(5).
Sedjo, R.A. 1990. The global carbon cycle. Journal of Forestry 88(10):
33-34.
Selby, M.J. 1982. Hillslope materials and processes. Oxford, U.K.: Oxford
University Press; 264 p.
Senft, R.L.; Rittenhouse, L.R.; Woodmansee, R.G. 1985. Factors
influencing patterns of cattle grazing behavior on shortgrass steppe.
Journal of Range Management 38(1): 82-87.
Servizi, J.A.; Martens, D.W. 1991. Effect of temperature, season, and
fish size on acute lethality of suspended sediments to coho salmon
(Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic
Sciences 48(3): 493-497.
Settergren, C.D. 1977. Impacts of river recreation use on streambank
soils and vegetation state-of-the-knowledge. In: Proceedings of the
river recreation management and research symposium. Gen. Tech.
Rep. NC-28. St. Paul, MN: North Central Forest Experiment Station,
Forest Service, U.S. Department of Agriculture; 55-60.
Shapley, S.; Phillip, S.; Bishop, D.M. 1965. Sedimentation in a salmon
stream. Journal of the Fisheries Research Board of Canada 22(4): 919-
928.
Sharom, M.S.; Miles, J.R.W.; Harris, C.R.; McEwen, F.L. 1980. Persistence
of 12 insecticides in water. Water Research 14: 1089-1093.
Sharpley, A.N. 1986. The role of sediment in accelerated
eutrophication. In: Proceedings of the fourth federal interagency
sedimentation conference; 1986 March 24-27; Las Vegas, NV.
Interagency Advisory Committee on Water Data, Subcommittee on
Sedimentation; 8-37 to 8-47.
Shaw, P.A.; Maga, J.A. 1943. The effect of mining silt on yield of fry
from salmon spawning beds. California Fish and Game 29(1): 29-41.
114
Shelby, B.; Vaske, J.J.; Harris, R. 1988. User standards for ecological
impacts at wilderness campsites. Journal of Leisure Research 20(3):
245-256.
Sheridan, W.L.; McNeil, W.J. 1968. Some effects of logging on two
salmon streams in Alaska. Journal of Forestry 66(2): 128-133.
Short, D.A. 1987. Responses of a northern California coastal drainage
basin to land use disturbance. In: Erosion and sedimentation in the
Pacific Rim. Publ. no. 165. International Association of Hydrological
Sciences; 503-504.
Shupe, J.B.; Brotherson, J.D. 1985. Differential effects of cattle and
sheep grazing on high mountain meadows in the Strawberry Valley
of central Utah. Great Basin Naturalist 45(1): 141-149.
Sidle, R.C.; Drlica, D.M. 1981. Soil compaction from logging with a
low-ground pressure skidder in the Oregon Coast Ranges. Soil
Science Society of America Journal 45(6): 1219-1224.
Siewert, H.F.; Miller, C.J.; Torke, B.G. 1989. Water quality and
macroinvertebrate populations before and after a hazardous waste
cleanup. Water Resources Bulletin 25(3): 685-690.
Simons, D.B.; Li, R.M.; Anderson, B.A. 1981. Procedural guide for
estimating water and sediment yield from forest roads. Report
CER81-82DBS-RML-BAA. Region 5. San Francisco, CA: U.S.
Department of Agriculture, Forest Service.
Sing, E.F. 1986. River sediment budget. In: Proceedings of the fourth
federal interagency sedimentation conference; 1986 March 24-27; Las
Vegas, NV. Interagency Advisory Committee on Water Data,
Subcommittee on Sedimentation; 3-73 to 3-82.
Skinner, Q.D.; Adams, J.C.; Rechard, P.A.; Beetle, A.A. 1974. Effect of
summer use of a mountain watershed on bacterial water quality.
Journal of Environmental Quality 3: 329-335.
Smart, G. 1984. Sediment transport formula for steep channels. Journal
of Hydraulic Engineering 110(3): 267-276.
Smith, D.G. 1976. Effect of vegetation on lateral migration of a
glacial meltwater river. Geological Society of America Bulletin
87(6): 857-860.
Smith, R.A.; Alexander, R.B.; Wolman, M.G. 1987. Water quality trends
in the Nation’s rivers. Science 235: 1607-1615.
Smith, R.B.; Commandeur, P.R.; Ryan, M.W. 1984. Vegetative succession,
soil development and forest productivity on landslides, Queen
Charlotte Islands, British Columbia, Canada. In: O’Loughlin, C.L.;
Pearce, A.J., eds. Symposium on effects of forest land use on erosion
and slope stability. Honolulu, HI: University of Hawaii; 109-116.
Smith, R.D.; Swanson, F.J. 1987. Sediment routing in a small drainage
basin in the blast zone at Mount St. Helens, Washington, USA.
Geomorphology 1:1-13.
Snider, M.D.; Miller, R.F. 1985. Effects of tractor logging on soils and
vegetation in Eastern Oregon. Soil Science Society of America Journal
49(5): 1280-1282.
Sollins, P.; McCorison, F.M. 1981. Nitrogen and carbon solution chemistry
of an old-growth coniferous watershed before and after cutting.
Water Resources Research 17(5): 1409-1418.
Sonntag, N.C.; Everitt, R.R.; Rattie, L.P.; Colnett, D.L.; Wolf, C.P.; Truett,
J.C.; Dorcey, A.H.J.; Holling, C.S. 1987. Cumulative effects
assessment: a context for further research and development. Canadian
Environmental Assessment Research Council; 96 p.
Sophocleous, M.; Townsend, M.A.; Whittemore, D.O. 1990. Movement
and fate of atrazine and bromide in central Kansas croplands.
Journal of Hydrology 115: 115-138.
Springer, E.P.; Coltharp, G.B. 1980. Effects of logging roads on storm
hydrographs. In: Symposium on watershed management 1980. New
York: American Society of Civil Engineers; 228-239.
Standiford, R.B.; Ramacher, S.I., eds. 1981. Cumulative effects of forest
management on California watersheds. Proceedings of the Edgebrook
Conference; 1980 June 2-3; Berkeley, CA. Publ. no. 3268. University
of California Division of Agricultural Science; 109 p.
Stankowski, S.J. 1972. Population density as an indirect indicator of
urban and suburban land-surface modifications. Prof. Paper 800-B.
Washington, DC: Geological Survey, U.S. Department of Interior;
219-224.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Stark, J.M.; Redente, E.F. 1985. Soil-plant diversity relationships on a
disturbed site in northwestern Colorado. Soil Science Society of
America Journal 49(4): 1028-1034.
Starr, P.; Hilborn, R. 1988. Reconstruction of harvest rates and stock
contribution in gauntlet salmon fisheries: application to British
Columbia and Washington sockeye (Oncorhynchus nerka). Canadian
Journal of Fisheries and Aquatic Sciences 45(12): 2216-2229.
Staver, K.W.; Brinsfield, R.B. 1990. Patterns of soil nitrate availability
in corn production systems: implications for reducing
groundwater contamination. Journal of Soil and Water Conservation
45(2): 318-323.
Steen, A.E.; Schubel, J.R. 1986. An application of a strategy to reduce
entrainment mortality. Journal of Environmental Management 23(3):
215-228.
Steinbrenner, E.C.; Gessel, S.P. 1955. The effect of tractor logging on
physical properties of some forest soils in southwestern Washington.
Soil Science Society of America Proceedings 19:372-376.
Stephenson, G.R.; Rychert, R.C. 1982. Bottom sediment: a reservoir of
Escherichia coli in rangeland streams. Journal of Range Management
35(1): 119-123.
Stephenson, G.R.; Veigel, A. 1987. Recovery of compacted soil on
pastures used for winter cattle feeding. Journal of Range Management
40(1): 46-49.
Stewart, W.P.; Carpenter, E.H. 1989. Solitude at Grand Canyon: an
application of expectancy theory. Journal of Leisure Research 21(1):
4-17.
Stott, T.A.; Ferguson, R.I.; Johnson, R.C.; Newson, M.D. 1986. Sediment
budgets in forested and unforested basins in upland Scotland. In:
Drainage basin sediment delivery. Publ. no. 159. International
Association of Hydrological Sciences; 57-68.
Stowell, R.; Espinosa, A.; Bjornn, T.C.; Platts, W.S.; Burns, D.C.; Irving,
J.S. 1983. Guide to predicting salmonid response to sediment yields
in Idaho Batholith watersheds. Regions 1, 4. U.S. Department of
Agriculture, Forest Service.
Stransky, J.J. 1981. Site preparation effects on soil bulk density and
pine seedling growth. Southern Journal of Applied Forestry 5(4):
176-180.
Striffler, W.D. 1959. Effects of forest cover on soil freezing in northern
lower Michigan. Station Paper 76. St. Paul, MN: Lake States Forest
Experiment Station, Forest Service, U.S. Department of Agriculture;
16 p.
Strothmann, R.O. 1979. Regeneration of Douglas-fir cutblocks on the
Six Rivers National Forest in northwestern California. Res. Paper
PSW-142. Berkeley, CA: Pacific Southwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 9 p.
Sturges, D.L. 1983. Long-term effects of big sagebrush control on
vegetation and soil water. Journal of Range Management 36(6):
760-765.
Sullivan, K. 1985. Long-term patterns of water quality in a managed
watershed in Oregon: 1. Suspended sediment. Water Resources
Bulletin 21(6): 977-987.
Sullivan, K.; Tooley, J.; Doughty, K.; Caldwell, J.E.; Knudsen, P. 1990.
Evaluation of prediction models and characterization of stream
temperature regimes in Washington. Timber/Fish/Wildlife Report
no. TFW-WQ3-90-006. Olympia, WA: Washington Department of
Natural Resources; 224 p.
Summer, R.M. 1980. Impact of horse traffic on trails in Rocky Mountain
National Park. Journal of Soil and Water Conservation 35:85-87.
Sutherland, R.A.; Bryan, R.B. 1991. Sediment budgeting: a case study in
the Katiorin drainage basin, Kenya. Earth Surface Processes and
Landforms 16(5): 383-398.
Swales, S.; Levings, C.D. 1989. Role of off-channel ponds in the life
cycle of coho salmon (Oncorhynchus kisutch) and other juvenile
salmonids in the Coldwater River, British Columbia. Canadian
Journal of Fisheries and Aquatic Sciences 46(2): 232-242.
Swank, W.T.; Helvey, J.D. 1970. Reduction of streamflow increases
following regrowth of clearcut hardwood forests. Publ. no. 96.
International Association of Scientific Hydrology; 346-359.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Swanson, F.J.; Dyrness, C.T. 1975. Impact of clearcutting and road
construction on soil erosion by landslides in the western Cascade
Range, Oregon. Geology 3: 393-396.
Swanson, F.J.; Fredriksen, R.L. 1982. Sediment routing and budgets:
implications for judging impacts of forestry practices. In: Sediment
budgets and routing in forest catchments. Gen. Tech. Rep. PNW-141.
Portland, OR: Pacific Northwest Forest and Range Experiment Station,
Forest Service, U.S. Department of Agriculture; 129-137.
Swanson, F.J.; Fredriksen, R.L.; McCorison, F.M. 1982. Material transfer
in a western Oregon forested watershed. In: Edmonds, R.L., ed.
Analysis of coniferous forest ecosystems in the Western United States.
Stroudsburg, PA: Hutchinson Ross Publication Co.; 233-266.
Swanson, F.J.; Lienkaemper, G.W. 1978. Physical consequences of large
organic debris in Pacific Northwest streams. Gen. Tech. Rep. PNW-
69. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 12 p.
Swanston, D.N. 1967. Soil water piezometry in a southeast Alaska
landslide area. Res. Note PNW-68. Portland, OR: Pacific Northwest
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture; 17 p.
Swanston, D.N. 1969. Mass wasting in coastal Alaska. Res. Paper PNW-
83. Portland, OR: Pacific Northwest Forest and Range Experiment
Station, Forest Service, U.S. Department of Agriculture; 15 p.
Swanston, D.N. 1970. Mechanics of debris avalanching in shallow till
soils of southeast Alaska. Res. Paper PNW-103. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 17 p.
Swanston, D.N.; Marion, D.A. 1991. Landslide response to timber harvest
in southeast Alaska. In: Proceedings of the fifth federal interagency
sedimentation conference; 1991 March 18-21; Las Vegas, NV. Federal
Energy Regulatory Commission; 10-49 to 10-56.
Swanston, D.N.; Swanson, F.J. 1976. Timber harvesting, mass erosion,
and steepland forest geomorphology in the Pacific Northwest. In:
Coates, D.R., ed. Geomorphology and engineering. Strousbury, PA:
Dowden, Hutchinson and Ross; 221.
Swift, L.W., Jr. 1984a. Gravel and grass surfacing reduces soil loss
from mountain roads. Forest Science 30(3): 657-670.
Swift, L.W., Jr. 1984b. Soil losses from roadbeds and cut and fill slopes
in the southern Appalachian Mountains. Southern Journal of Applied
Forestry 8(4): 209-215.
Tarrant, R.F. 1956. Effect of slash burning on some physical soil
properties. Forest Science 2:18-22.
Taylor, E.B. 1988. Water temperature and velocity as determinants of
microhabitats of juvenile chinook and coho salmon in a laboratory
stream channel. Transactions of the American Fisheries Society 117(1):
22-28.
Taylor, H.M.; Roberson, G.M.; Parker, J.J., Jr. 1966. Soil strength - root
penetration relations for medium- to coarse-textured soil materials.
Soil Science 102:18-22.
Temmink, J.H.M.; Field, J.A.; van Haastrecht, J.C.; Merkelbach, R.C.M.
1989. Acute and sub-acute toxicity of bark tannins in carp (Cyprinus
carpio L.). Water Research 23(3): 341-344.
Temple, D.M. 1982. Flow retardance of submerged grass channel linings.
Transactions of the American Society of Agricultural Engineers 25(5):
1300-1303.
Thene, J.R.; Stefan, H.G.; Daniil, E.I. 1989. Low-head hydropower
impacts on stream dissolved oxygen. Water Resources Bulletin 25(6):
1189-1198.
Theurer, F.D.; Lines, I.; Nelson, T. 1985. Interaction between riparian
vegetation, water temperature, and salmonid habitat in the Tucannon
River. Water Resources Bulletin 21(1): 53-64.
Thomas, R.B. 1990. Problems in determining the return of a watershed
to pretreatment conditions: techniques applied to a study at Caspar
Creek, California. Water Resources Research 26(9): 2079-2088.
Thomas, R.E.; Gharrett, J.A.; Carls, M.G.; Rice, S.D.; Moles, A.; Korn, S.
1986. Effects of fluctuating temperature on mortality, stress, and
energy reserves of juvenile coho salmon. Transactions of the American
Fisheries Society 115(1): 52-59.
115
Thompson, M.A. 1990. Determining impact significance in EIA: a
review of 24 methodologies. Journal of Environmental Management
30(3): 235-250.
Thorne, R.E.; Ames, J.J. 1987. A note on variability of marine survival
of sockeye salmon (Oncorhynchus nerka) and effects of flooding on
spawning success. Canadian Journal of Fisheries and Aquatic Sciences
44(10): 1791-1795.
Thornthwaite, C.W.; Mather, J.R. 1955. The water balance. Publ. no. 8.
Centerton, NJ: Laboratory of Climatology.
Thornthwaite, C.W.; Mather, J.R. 1957. Instructions and tables for
computing potential evapotranspiration and the water balance.
Publ. no. 18. Centerton, NJ: Laboratory of Climatology.
Thurow, T.L.; Blackburn, W.H.; Taylor, C.A. 1988. Infiltration and
interrill erosion responses to selected livestock grazing strategies,
Edwards Plateau, Texas. Journal of Range Management 41(4):
296-302.
Tiedemann, A.R.; Higgins, D.A.; Quigley, T.M.; Sanderson, H.R. 1989.
Stream chemistry responses to 4 range management strategies in
eastern Oregon. Res. Paper PNW-413. Portland, OR: Pacific Northwest
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture; 9 p.
Tiedemann, A.R.; Higgins, D.A.; Quigley, T.M.; Sanderson, H.R.; Marx,
D.B. 1987. Responses of fecal coliform in streamwater to four
grazing strategies. Journal of Range Management 40(4): 322-329.
Tiedemann, A.R.; Quigley, T.M.; Anderson, T.D. 1988. Effects of timber
harvest on stream chemistry and dissolved nutrient losses in
northeast Oregon. Forest Science 34(2): 344-358.
Tim, U.S.; Mostaghimi, S. 1989. Model for predicting transport of
pesticides and their metabolites in the unsaturated zone. Water
Resources Bulletin 25(4): 765-774.
Toews, D.A.A.; Moore, M.K. 1982. The effects of streamside logging on
large organic debris in Carnation Creek. British Columbia Ministry
of Forests and Land Management Report 11; 29 p.
Tollner, E.W.; Barfield, B.J.; Haan, C.T.; Kao, T.Y. 1976. Suspended
sediment filtration capacity of rigid vegetation. Transactions of the
American Society of Agricultural Engineers 19(4): 678-682.
Touysinhthiphonexay, K.C.N.; Gardner, T.W. 1984. Threshold response
of small streams to surface coal mining, bituminous coal fields,
Central Pennsylvania. Earth Surface Processes and Landforms 9(1):
43-58.
Towne, G.; Owensby, C. 1984. Long-term effects of annual burning at
different dates in ungrazed Kansas tallgrass prairie. Journal of
Range Management 37(5): 392-397.
Toy, T.J. 1984. Geomorphology of surface-mined lands in the western
United States. In: Costa, J.E.; Fleisher, P.J., eds. Developments and
applications of geomorphology. Berlin: Springer-Verlag; 133-170.
Treshow, M.; Stewart, D. 1973. Ozone sensitivity of plants in natural
communities. Biological Conservation 5(3): 209-214.
Trimble, G.R.; Weitzman, S. 1953. Soil erosion on logging roads. Soil
Science Society of America Proceedings 17: 152-154.
Trimble, S.W. 1977. The fallacy of stream equilibrium in contemporary
denudation studies. American Journal of Science 277: 876-887.
Trimble, S.W. 1983. A sediment budget for Coon Creek basin in the
driftless area, Wisconsin, 1853-1977. American Journal of Science
283: 454-474.
Triska, F.J.; Kennedy, V.C.; Avanzino, R.J.; Zellweger, G.W.; Bencala,
K.E. 1989. Retention and transport of nutrients in a third-order
stream: channel processes. Ecology 70(6): 1877-1892.
Troendle, C.A. 1970. The flow interval method for analyzing timber
harvesting effects on streamflow regimen. Water Resources Research
6(1): 328-332.
Troendle, C.A. 1988. Effect of partial cutting and thinning on the water
balance of the subalpine forest. In: Proceedings, future forests of the
Mountain West: a stand culture symposium; 1986 Sept 29 - Oct 3;
Missoula, MT. Gen. Tech. Rep. INT-243. Ogden, UT: Intermountain
Forest and Range Experiment Station, Forest Service, U.S. Department
of Agriculture.
116
Truhlar, J.F.; Reed, L.A. 1976. Occurrence of pesticide residues in four
streams draining different land-use areas in Pennsylvania, 1969-
1971. Pesticide Monitoring Journal 10(3): 101-110.
Trustrum, N.A.; DeRose, R.C. 1988. Soil depth-age relationship of
landslides on deforested hillslopes, Taranaki, New Zealand.
Geomorphology 1(2): 143-160.
Tschaplinski, P.J.; Hartman, G.F. 1983. Winter distribution of juvenile
coho salmon (Oncorhynchus kisutch), before and after logging in
Carnation Creek, British Columbia, and some implications for
overwinter survival. Canadian Journal of Fisheries and Aquatic Sciences
40(4): 452-461.
United States Forest Service. 1974. Forest hydrology part II - Hydrologic
effects of vegetation manipulation. Missoula, MT: U.S. Department
of Agriculture, Forest Service; 229 p.
United States Forest Service. 1980. An approach to water resources
evaluation of non-point silvicultural sources (a procedural
handbook). EPA-600/8-80-012. Athens, GA: Environmental Research
Laboratory; 864 p.
United States Forest Service. 1988. Cumulative off-site watershed effects
analysis. In: USFS Region 5 Soil and Water Conservation Handbook.
FSH 2509.22. Region 5. San Francisco, CA: U.S. Department of
Agriculture, Forest Service.
Ursic, S.J.; Esher, R.J. 1988. Influence of small mammals on stormflow
responses of pine-covered catchments. Water Resources Bulletin
24(1): 133-139.
Van Cleve, K.; Moore, T.A. 1978. Cumulative effects of nitrogen,
phosphorous, and potassium fertilizer additions on soil respiration,
pH and organic matter content. Soil Science Society of America
Journal 42(1): 121-124.
Van Donsal, D.J.; Geldreich, E.E. 1971. Relationships of salmonellae to
fecal coliforms in bottom sediments. Water Research 5:1079-1087.
Van Hylckama, T.E.A. 1974. Water use by saltcedar as measured by the
water budget method. Prof. Paper 491-E. Washington, DC: U.S.
Department of Interior, Geological Survey; 30 p.
Van Kooten, G.C.; Arthur, L.M. 1989. Assessing economic benefits of
climate change on Canada’s boreal forest. Canadian Journal of
Forest Research 19: 463-470.
Van Nieuwenhuyse, E.E.; LaPerriere, J.D. 1986. Effects of placer gold
mining on primary production in subarctic streams of Alaska.
Water Resources Bulletin 22(1): 91-99.
Van Poollen, H.W.; Lacey, J.R. 1978. Herbage response to grazing
systems and stocking intensities. Journal of Range Management 32(4):
250-253.
Van Sickle, J. 1981. Long-term distributions of annual sediment yields
from small watersheds. Water Resources Research 17(3): 659-663.
Van Sickle, J.; Gregory, S.V. 1990. Modeling inputs of large woody
debris to streams from falling trees. Canadian Journal of Forest
Resources 20(10): 1593-1601.
Vanoni, V.A., ed. 1975. Sedimentation engineering. New York: American
Society of Civil Engineers; 745 p.
Veblen, T.T.; Lorenz, D.C. 1986. Anthropogenic disturbance and recovery
patterns in montane forests, Colorado Front Range. Physical
Geography 7(1): 1-24.
Vice, R.B.; Ferguson, G.E.; Guy, H.P. 1968. Erosion from suburban
highway construction. Journal of the Hydraulics Division, American
Society of Civil Engineers 94(HY1): 347-348.
Vitousek, P.M.; Matson, P.A. 1985. Intensive harvesting and site
preparation decrease soil nitrogen availability in young plantations.
Southern Journal of Applied Forestry 9(2): 120-125.
Vitousek, P.M.; Melillo, J.M. 1979. Nitrate losses from disturbed forests:
patterns and mechanisms. Forest Science 25: 605-619.
Voorhees, W.B. 1983. Relative effectiveness of tillage and natural forces
in alleviating wheel-induced soil compaction. Soil Science Society
of America Journal 47(1): 129-133.
Voorhees, W.B.; Young, R.A.; Lyles, L. 1979. Wheel traffic considerations
in erosion research. Transactions of the American Society of
Agricultural Engineers 22: 786-790.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
Von Bertalanffy, L. 1968. General systems theory. New York: George
Braziller, Inc.
Waide, J.B.; Swank, W.T. 1976. Nutrient recycling and the stability of
ecosystems: implications for forest management in the southeastern
U.S. In: Proceedings of the 1975 national convention of the Society of
American Foresters; 1976 Sept 28 - Oct 2; Washington, DC; 404-424.
Walker, D.A.; Everett, K.R. 1987. Road dust and its environmental
impact on Alaskan taiga and tundra. Arctic and Alpine Research
19(4): 479-489.
Walker, J.W.; Heitschmidt, R.K. 1986. Effect of various grazing systems
on type and density of cattle trails. Journal of Range Management
39(5): 428-430.
Wall, G.; Wright, C. 1977. The environmental impact of outdoor
recreation. Publ. no. 11. Waterloo, ON: University of Waterloo,
Department of Geography.
Walling, D.E.; Gregory, K.J. 1970. The measurement of the effects of
building construction on drainage basin dynamics. Journal of
Hydrology 11:129-144.
Walsh, R.G.; Bjonback, R.D; Aiken, R.A.; Rosenthal, D.H. 1990. Estimating
the public benefits of protecting forest quality. Journal of
Environmental Management 30(2): 175-189.
Walter, R.; Macht, W.; Durkop, J.; Hecht, R.; Hornig, U.; Schulze, P.
1989. Virus levels in river waters. Water Research 23(2): 133-138.
Warren, S.D.; Blackburn, W.H.; Taylor, C.A., Jr. 1986a. Effects of season
and stage of rotation cycle on hydrologic condition of rangeland
under intensive rotation grazing. Journal of Range Management
39(6): 486-490.
Warren, S.D.; Blackburn, W.H.; Taylor, C.A., Jr. 1986b. Soil hydrologic
response to number of pastures and stocking density under intensive
rotation grazing. Journal of Range Management 39(6): 500-504.
Warren, S.D.; Thurow, T.L.; Blackburn, W.H.; Garza, N.E. 1986c. The
influence of livestock trampling under intensive rotation grazing
on soil hydrologic characteristics. Journal of Range Management
39(6): 491-495.
Watson, C.C.; Harvey, M.D.; Garbrecht, J. 1986. Geomorphic-hydraulic
simulation of channel evolution. In: Proceedings of the fourth federal
interagency sedimentation conference; 1986 March 24-27; Las Vegas,
NV. Interagency Advisory Committee on Water Data, Subcommittee
on Sedimentation; 5-21 to 5-30.
Watts, F.J.; Burroughs, E.R., Jr.; King, J.G.; Hansen, D. 1986. Evaluation
of a road sediment model. In: Proceedings of the fourth federal
interagency sedimentation conference; 1986 March 24-27; Las Vegas,
NV. Interagency Advisory Committee on Water Data, Subcommittee
on Sedimentation; 6-171 to 6-180.
Weaver, T.; Dale, D. 1978. Trampling effects of hikers, motorcycles
and horses in meadows and forests. Journal of Applied Ecology 15:
451-457.
Wehrhahn, C.F.; Powell, R. 1987. Electrophoretic variation, regional
differences, and gene flow in coho salmon (Oncorhynchus kisutch)
of southern British Columbia. Canadian Journal of Fisheries and
Aquatic Sciences 44: 822-831.
Weidman, R.H. 1920. A study of windfall loss of western yellow pine in
selection cuts 15-30 years old. Journal of Forestry 18: 616-622.
Welch, E.B.; Horner, R.R.; Patmont, C.R. 1989. Prediction of nuisance
periphytic biomass: a management approach. Water Research 23(4):
401-405.
Wendler, H.O.; Deschamps, G. 1955. Logging dams on coastal Washington
streams. 1. Olympia, WA: Washington Dept. of Fisheries; 27-38.
Wert, S.; Thomas, B.R. 1981. Effects of skid roads on diameter, height
and volume growth in Douglas-fir. Soil Science Society of America
Journal 45: 629-632.
West, N.E.; Provenza, F.D.; Johnson, P.S.; Owens, M.K. 1984. Vegetation
change after 13 years of livestock grazing on sagebrush semidesert
in west central Utah. Journal of Range Management 37(3): 262-264.
Wester, L. 1981. Composition of native grasslands in the San Joaquin
Valley, California. Madrono 28(4): 231-241.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.
White, W.R.; Milli, H.; Crabbe, A.D. 1978. Sediment transport; an
appraisal of available methods. Publ. no. 119. Wallingford, UK:
United Kingdom Hydraulics Research Station.
Wight, J.R.; Godfrey, E.B. 1985. Predicting yield response to nitrogen
fertilization on northern Great Plains rangelands. Journal of Range
Management 38(3): 238-241.
Willard, B.E.; Marr, J.W. 1970. Effects of human activities on alpine
tundra ecosystems in Rocky Mountain National Park, Colorado.
Biological Conservation 2(4): 257-265.
Willard, B.E.; Marr, J.W. 1971. Recovery of alpine tundra under
protection after damage by human activities in the Rocky Mountains
of Colorado. Biological Conservation 3(3): 181-190.
Williams, G.P. 1983. Improper use of regression equations in earth
sciences. Geology 11: 195-197.
Williams, G.P.; Wolman, M.G. 1984. Downstream effects of dams on
alluvial rivers. Prof. Paper 1286. Washington, DC: Geological Survey,
U.S. Department of Interior; 83 p.
Williams, J.R. 1975. Sediment routing for agricultural watersheds.
Water Resources Bulletin 11(5): 965-974.
Wilson, A.A.G.; Ferguson, I.S. 1986. Predicting the probability of house
survival during bushfires. Journal of Environmental Management
23(3): 259-270.
Wilzbach, M.A.; Cummins, K.W.; Hall, J.D. 1986. Influence of habitat
manipulations on interactions between cutthroat trout and
invertebrate drift. Ecology 67(4): 898-911.
Wischmeier, W.H. 1976. Use and misuse of the Universal Soil Loss
Equation. Journal of Soil and Water Conservation 31: 5-9.
Wischmeier, W.H.; Smith, D.D. 1978. Predicting rainfall erosion losses—
a guide to conservation planning. Handbook 537. Washington, DC:
U.S. Department of Agriculture; 58 p.
Wissmar, R.C.; Simenstad. C.A. 1988. Energetic constraints of juvenile
chum salmon (Oncorhynchus keta) migrating in estuaries. Canadian
Journal of Fisheries and Aquatic Sciences 45(9): 1555-1560.
Wolfe, M.D.; Williams, J.W. 1986. Rates of landsliding as impacted by
timber management activities in northwestern California. Bulletin
of the Association of Engineering Geologists 23(1): 53-60.
Wolman, M.G.; Brush, L.M., Jr. 1961. Factors controlling the size and
shape of stream channels in coarse noncohesive sands. Prof. Paper
282-G. Washington, DC: U.S. Department of Interior, Geological Survey;
183-210.
Wolman, M.G.; Schick, A.P. 1967. Effects of construction on fluvial
sediment, urban and suburban areas of Maryland. Water Resources
Research 3: 451-464.
Wood, C.C. 1987a. Predation of juvenile Pacific salmon by the common
merganser (Mergus merganser) on eastern Vancouver Island I:
Predation during the seaward migration. Canadian Journal of Fisheries
and Aquatic Sciences 44(5): 941-949.
Wood, C.C. 1987b. Predation of juvenile Pacific salmon by the common
merganser (Mergus merganser) on eastern Vancouver Island II:
Predation of stream-resident juvenile salmon by merganser broods.
Canadian Journal of Fisheries and Aquatic Sciences 44(5): 950-959.
Wood, J.C.; Blackburn, W.H.; Pearson, H.A.; Hunter, T.K. 1989. Infiltration
and runoff water quality response to silvicultural and grazing
treatments on a longleaf pine forest. Journal of Range Management
42(5): 378-389.
Wood, M.K.; Blackburn, W.H. 1984. Vegetation and soil responses to
cattle grazing systems in the Texas rolling plains. Journal of Range
Management 37(4): 303-308.
Woods, L.E.; Schuman, G.E. 1988. Cultivation and slope position effects
on soil organic matter. Soil Science Society of America Journal
52(5): 1371-1376.
Woodward, D.F. 1978. Assessing the hazard of picloram to cutthroat
trout. Journal of Range Management 32(3): 230-231.
Woodward, D.F. 1982. Acute toxicity of mixtures of range management
herbicides to cutthroat trout. Journal of Range Management 35(4):
539-540.
117
Wright, K.A.; Sendek, K.H.; Rice, R.M.; Thomas, R.B. 1990. Logging
effects on streamflow: storm runoff at Caspar Creek in northwestern
California. Water Resources Research 26(7): 1657-1668.
Wu, T.H.; Swanston, D.N. 1980. Risk of landslides in shallow soils and
its relation to clearcutting in southeastern Alaska. Forest Science
26(3): 495-510.
Yee, C.S.; Harr, R.D. 1977. Effect of wetting mode on shear strength of
two aggregated soils. Res. Note PNW-303. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 9 p.
Yin, X.; Perry, J.A.; Dixon, R.K. 1989. Influence of canopy removal on
oak forest floor decomposition. Canadian Journal of Forest Research
19(2): 204-214.
Yost, C.; Naney, J.W. 1975. Earth-dam seepage and related land and
water problems. Journal of Soil and Water Conservation 30: 87-91.
Yousef, Y.A.; McLellan, W.M.; Zebuth, H.H. 1980. Changes in phosphorus
concentrations due to mixing by motorboats in shallow lakes. Water
Research 14: 841-852.
Ziemer, R.R. 1964. Summer evapotranspiration trends as related to
time after logging of forests in Sierra Nevada. Journal of Geophysical
Research 69(4): 615-620.
Ziemer, R.R. 1979. Evaporation and transpiration. Reviews of Geophysics
and Space Physics 17(6): 1175-1186.
Ziemer, R.R. 1981. Storm flow response to road building and partial
cutting in small streams of northern California. Water Resources
Research 17(4): 907-917.
118
Ziemer, R.R.; Lewis, J.; Rice, R.M.; Lisle, T.E. 1991. Modeling the
cumulative watershed effects of forest management strategies. Journal
of Environmental Quality 20(1): 36-42.
Ziemer, R.R.; Swanston, D.N. 1977. Root strength changes after logging
in southeast Alaska. Res. Note PNW-306. Portland, OR: Pacific
Northwest Forest and Range Experiment Station, Forest Service, U.S.
Department of Agriculture; 10 p.
Zimmerman, G.T.; Neuenschwander, L.F. 1983. Fuel-load reductions
resulting from prescribed burning in grazed and ungrazed Douglas-
fir stands. Journal of Range Management 36(3): 346-350.
Zimmerman, G.T.; Neuenschwander, L.F. 1984. Livestock grazing
influences on community structure, fire intensity, and fire frequency
within the Douglas-fir/Ninebark habitat type. Journal of Range
Management 37(2): 104-110.
Zimmerman, R.C.; Goodlett, J.C.; Comer, G.H. 1967. The influence of
vegetation on channel form of small streams. In: Symposium on
river morphology. Publ. no. 75. International Association of Scientific
Hydrology; 255-275.
Zobel, D.B.; Lewis, L.F.; Hawk, G.M. 1985. Ecology, pathology, and
management of Port-Orford-cedar (Chamaecyparis lawsoniana).
Gen. Tech. Rep. PNW-184. Portland, OR: Pacific Northwest Forest
and Range Experiment Station, Forest Service, U.S. Department of
Agriculture; 161 p.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-141-WWW. 1993.